ML20128M201
| ML20128M201 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 10/09/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20128M120 | List: |
| References | |
| 50-245-96-06, 50-245-96-6, 50-336-96-06, 50-336-96-6, 50-423-96-06, 50-423-96-6, NUDOCS 9610160023 | |
| Download: ML20128M201 (70) | |
See also: IR 05000245/1996006
Text
. - -
- _
._
-
__ -
-
-, . . - - . ..
_ _ .
_ _ . _ . _ - -
.
'
.
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
Docket Nos.:
50-245
50-336
50-423
Report Nos.:
96-06
96-06
96-06
License Nos.:
DPR-65
Licensee:
Northeast Nuclear Energy Company
P. O. Box 128
Waterford, CT 06385
Facility:
Millstone Nuclear Power Station, Units 1,2, and 3
,
inspection at:
Waterford, CT
Dates:
June 27,1996 - August 26,1996
,
Inspectors:
T. A. Easlick, Senior Resident inspector Unit 1
P. D. Swetland, Senior Resident inspector, Unit 2
A. C. Cerne, Senior Resident inspector, Unit 3
A. L. Burritt, Resident inspector, Unit 1
4
D. P. Beaulieu, Resident inspector, Unit 2
'
R. J. Arrighi, Resident inspector, Unit 3
J. T. Shediosky, Senior Reactor Analyst
'
J. M. Trapp, Senior Reactor A.ialyst
J. T. Furia, Senior Radiation Specialist
1
P. M. Peterson, NDE Technician
J. H. Lusher, Health Physicist
G. C. Smith, Senior Physical Security inspector
i
Approved by:
Jacque P. Durr, Chief
Projects Branch No. 6
Division of Reactor Projects
j
.
9610160023 961009
ADOCK 05000245
G
_
_ _
_
_ _ _ _ . . _
.
.
TABLE OF CONTENTS
EXECUTIVE SU M M ARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iv
...........
U1.1 Operations
1
..................................................
U1.01
Conduct of Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
U1.08
Miscellaneous Operations issues (92700)
3
..............
U 1.ll M aintena nce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
U1.M1
Conduct of Maintenance
3
..........................
U 1.Ill Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
U1.E8
Miscellaneous Engineering issues . . . . . . . . . . . . . . . . . . . . 7
,
I
U 2.1 O pe r a ti on s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
U2.01
Conduct of Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
,
U2.03
Operations Procedures and Documentation
10
i
............
U2.07
Quality Assurance in Operations . . . . . . . . . . . . . . . . . . . . 12
U2.08
Miscellaneous Operations issues (92700)
12
.............
{
U 2. li M ain te n a n c e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
U2.M1
Conduct of Maintenance
13
.........................
U2.M8
Miscellaneous Maintenance issues
13
..................
U 2.lli Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
U2.E1
Conduct of Engineering
16
..........................
U2.E2
Engineering Support of Facilities and Equipment . . . . . . . . . 20
U2.E8
Miscellaneous Engineering Issues . . . . . . . . . . . . . . . . . . . 25
U3.1 Operations
33
.................................................
U3.01
Conduct of Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
U3.07
Quality Assurance in Operations . . . . . . . . . . . . . . . . . . . . 34
U3.08
Miscellaneous Operations issues (92700)
34
.............
U 3.Il M aintena n ce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
U3.M1
Conduct of Maintenance
36
.........................
U3.M3
Maintenance Procedures and Documentation . . . . . . . . . . . 37
U 3.Ill En gin e e rin g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
U3.E1
Conduct of Engineering
38
..........................
U3.E2
Engineering Support of Facilities and Equipment . . . . . . . . . 40
U3.E7
Quality Assurance in Engineering Activities . . . . . . . . . . . . 44
U3.E8
Miscellaneous Engineering issues . . . . . . . . . . . . . . . . . . . 45
3
IV Plant Support
46
.................................................
R1
Radiological Protection and Chemistry Controls
46
.........
ii
- _ -
_
_ . - - . .
--
.-- .
_ - ..
-
.
.
R2
Status of Radiological Protection and Chemistry
Facilities and Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . 48
R5
Staff Training and Qualification in Radiological
Protection & Chemistry
50
..........................
R8
Miscellaneous Radiological Protection & Chemistry
,
Issues.......................................
50
,
P3
EP Procedures and Documentation
51
..................
S1
Conduct of Security and Safeguards Activities . . . . . . . . . . 51
S8
Miscellaneous Security and Safeguards issues . . . . . . . . . . 52
V. Management Meetings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
X1
Exit Meeting Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
X3
Management Meeting Summary . . . . . . . . . . . . . . . . . . . . 55
r
I
l
.
iii
.-
-
-
_ _ _
_ _ _ _ _
__
_. . . _ _ _ _ _ _
_
._.
,
>
.
.
EXECUTIVE SUMMARY
Millstone Nuclear Power Station
Combined Inspection 245/96-06;336/96-06;423/96-06
Operations
-
A routine review of a Unit 1 operability determination (OD) was performed.
The ODs are used to assess degraded plant conditions which affect equipment
operability and provide compensatory measures as applicable. Licensed
operators were not cognizant of operability determinations that assessed and
i
,
dispositioned degraded plant conditions. Further, no process or control
existed to ensure that licensed operators review and remain cognizant of ODs,
including compensatory actions. The failure of licensed personnel to maintain
cognizance of degraded conditions which affect operability could adversely
impact the ability of the on-shift personnel to assess subsequent equipment
,
degradation. (Section U1.01.2)
On June 8,1996, a small electrical fire occurred in the Unit 1 drywell, which
lasted several minutes, and was extinguished with a dry chemical, handheld
fire extinguisher. Approximately four hours after the fire was extinguished, a
plant equipment operator noticed that the differential pressures across the
running standby gas treatment (SBGT) system train were high. The normal
ventilation system had been isolated and "B" SBGT was initiated to comply
4
with a technical specification action statement, following the radiation
monitoring systems being declared inoperable. After conferring with the Unit
Director and the Duty Officer, the Shift Manager elected to unisolate the
reactor building, start normal ventilation and secure the running train of SBGT,
in an effort to remove or " purge" the remaining dry chemical powder in the
d
drywell atmosphere. However, this action resulted in a deviation from the
requirements in Technical Specification 3.2.E.2. This issue is unresolved
pending further NRC review and inspection. (Section U1.01.3)
Unit 2 control of shutdown margin during plant cooldowns differs from the
'
design basis descriptions. Cold shutdown boron concentration is not attained
pnor to initiating cooldown nor is the shutdown group of control rods
" cocked" during the cooldown evolution. The failure to conduct safety
evaluations to support these deviations, and update the Final Safety Analysis
Report is considered an apparent violation. Also, the licensee's practice of
injecting unsampled volumes of boric acid into the reactor coolant system
during cooldowns was considered unresolved pending further esatuation of the
J
potential for inadvertent (RCS) dilution. (Section U2.03.1)
4
,
NRC review of 19 licensee event reports (LERs) found them to be generally
timely and informative. However, the number of LERs needing supplemental
information and the number of missed committed actions indicated
weaknesses in the licensee's program for development, review and tracking of
events. (Section U2.07.1)
.
iv
-
.
.
Routine operations of Unit 3 in cold shutdown (Mode 5) conditions were well
controlled, particularly with consideration of the appropriate shutdown risk
criteria. (Section U3.01.1)
An audit of the adequacy of Northeast Utilities quality assurance (QA) program
by the Joint Utility Management Assessment Team indicated that the audit,
surveillance, and inspection programs at Millstone were not effective in the
implementation of their Mission Statement and the resolution of identified
problems. The team attributed these problems to: lack of support for the QA
organization by executive and line management; and the lack of an effective
corrective action program. QA effectiveness is a restart issue and a part of
the Restart Assessment Plan. (Section U3.07.1)
The operation of Unit 3 outside of its design basis resulted from
nonconservative piping design and pipe support stress analyses. This
deficiency affected independent safety trains in multiple plant systems and is
considered an apparent violation. (Section U3.08.3)
Maintenance
On July 16,1996, plant personnel were removing a temporary filter assembly
from the Unit 1 spent fuel pool when a wire rope, attached to the filter
assembly, was entangled with control rods that were suspended from the
spent fuel pool equipment rail. This caused five control rods to shift position
away from the wall and come to rest against an adjacent spent fuel rack. Six
of the eight individuals involved in this evolution were contaminated as a
result the filter removal event. They were successfully decontaminated and
whole body counts indicated no internal dose was received. During the event,
no area radiation monitor alarms were received and no airborne radiation was
detected. Following an under water inspection, the suspended control rods
were stabilized with additional cables attached to the bottom of the rods and
secured to the refueling bridge. The licensee is currently developing a
recovery plan. This issue is unresolved pending further NRC review and
inspection. (Section U1.M1.1)
Three weaknesses were identified during the Unit 1 intergranular stress
corrosion cracking (lGSCC) program review. The IGSCC program lacks detail
to prevent inadvertent procedure oversights during ultrasonic testing (UT)
examinations and evaluations. The weaknesses are: no method to evaluate
unresolved UT indications, no specific UT procedure or calibration blocks for
the examinations, and no method to track or trend UT indications from outage
to outage. These weaknesses resulted in many UT indications being
incorrectly overturned and indications that were found in one inspection,
missed in the subsequent inspection. The method used to evaluate
Unresolved Indication Reports (UIR's) relies solely on Nondestructive (NDE)
Level 111 expertise. These weaknesses resulted in flawed components being
returned to service without performing an engineering evaluation. This is an
unresolved item. (Section U1.M1.2)
v
.
.
Unit 2 incorrectly changed the inservice testing program (IST) requirements for
high pressure safety injection (HPSI) pump discharge check valves. The
licensee appropriately determined that the IST backflow testing of these
valves could result in overloading a diesel generator under certain conditions.
However, the deferral of the quarterly test requirement to refueling intervals
was inappropriate because other mechanisms were available to safely conduct
these tests. The issue remained unresolved pending licensee actions to
correct the test regime. (Section U2.M8.2)
The pre-job brief for the loop calibration of the Unit 3 containment
recirculation pump flow transmitter was thorough. (Section U3.M1.1)
The licensee identified licensing basis discrepancies associated with a Unit 3
design change implementing the use of trisodium phosphate as a pH control
agent. These issues require resolution prior to plant heatup to mode 4.
(U3.M3.1)
Engineering
A review of a selected group of NCRs, for operability determinations, indicated
that the physical control of the NCR process was lost at Millstone Unit 1. The
NCR process appears to have been used as an identification process for
degraded and nonconforming conditions in the field, contrary to procedure
3.05. The failure of the licensee to properly utilize the NCR process in
accordance with procedure 3.05, written to comply with 10 CFR 50 Appendix
B Criterion XV, is considered an apparent violation. (Section U1.E8.1)
Unit 2 did not establish a uniform refueling boron concentration in the RCS
prior to securing RCPs. This was reasonable because they could not have
anticipated the need to perform a core off-load during this mid-cycle outage.
However, after identifying the need to off-load fuel in order to repair an
unisolable valve, licensee performance in dispositioning this problem was
weak in that they planned to drain the RCS to mid-loop when other options
involving less risk were available. In addition, PORC did not provide rigorous
oversight in approving a TS clarification that redefined " uniform" boron
concentration, such that, while meeting the intent of the TS, it would not
have complied with the TS as written. (Section U2.E1.1)
Unit 2 implementation of controls to reduce the potential for draining the
reactor cavity during refueling activities was inadequate. One potential non
seismic drain path was not isolated nor the consequences of its failure
formally evaluated and controlled. The issue is unresolved pending further
review of licensee commitments regarding this concern. (Section U2.E1.2)
Unit 2 investigation of potential water hammer events that damaged
emergency core cooling system suction piping supports was not timely or
comprehensive. More than a year after identification of support damage, the
root cause had not been found nor was a comprehensive assessment of
vi
.
.
system structural supports completed. The issue remains unresolved pending
fuither review of the cause and corrective actions. (Section U2.E8.1)
e
On November 15,1995, Unit 2 operated for eleven hours at a power level
slightb above the operating license requirement, due to erroneous steam
generator blowdown flow input into the plant computer calorimetric
calculation. This issue remained unresolved pending final licensee control over
blowdown flow input, and further NRC review of plant computer programming
i
controls. (Section U2.E8.2)
e
Unit 2 discovered that the containment sump screens had been incorrectly
constructed such that larger debris than analyzed could pass through to the
emergency core cooling systems (ECCSs). This potential common cause
failure of ECCSs is considered an apparent violation of the technical
specifications and corrective action program requirements. (Section U2.E8.4)
e
Unit 2 identified seven solenoid-operated valves (SOVs) inside containment
whose environmental qualification (EEQ) was incorrect. The valves, which
provide containment isolation for various post-accident monitoring and control
functions must close on a containment isolation signal, and then are reopened
to perform post-accident functions. The licensee had erroneously focused on
only the containment isolation functions and concluded that the valves fail-
safe. Therefore, EEQ of the SOV circuit was not required. In fact, the post-
accident fonction requires full EEQ of the circuits, and this qualification did not
exist for these seven valves. Further, the NRC determined that several other
weaknesses in the licensee's implementation of EEQ requirements raise
concern with the ability of other components to perform their functions in an
accident environment. This issue is considered an apparent violation.
(Section E2.E8.6)
e
The licensee continued to evaluate concerns at Units 2 and 3 related to
emergency core cooling system throttle valve flow restrictions and settings.
The potential for valve erosion to occur over a period of continuous
recirculation flow was considered. This problem, in conjunction with
regulatory guidance on containment sump screen sizing, requires the licensee
to implement a modification on Unit 3 to install orifice plates in the affected
lines. The conduct of licensee corrective measures for these problems is
considered an unresolved item pending NRC review for effectiveness.
(Sections U2.E8.4 and U3.E1.1)
e
The root cause investigations and corrective actions taken with regards to
selected level "A" and B" adverse condition reports was mixed. The review
identified examples of ineffective corrective actions and the failure of the
Events Analysis department to identify the discrepancies during their closecut
review. (Section U3.E2.1)-
j
i
vii
l
l
.
.
The licensee's failure to correctly translate the technical requirements of the
ASME Code, relating to the specification of replacement stud material for the
lower flange of the chemical volume and control system letdown heat
exchanger, into the design details of a plant design change record represents
an apparent violation of 10 CFR 50, Appendix' B. The premature closure of a
nonconformance report, written to track the heat exchanger leakage from the
lower flange, may have contributed to the licensee's lack of recognition of this
concern as a code compliance problem. (Section U3.E8.1)
Plant Support
The licensee continues to maintain an effective health physics program,
especially during outage operations. General plant radiological housekeeping
continues to significantly improve, especially in the Unit 1 turbine building.
The Radwaste Remediation Project at Unit 1 continues to make progress in
addressing the material condition deficiencies in the Unit 1 liquid waste
processing systems and facilities. Continued attention to work planning and
work control is necessary for improvement in maintaining occupational
exposures as low as is reasonably achievable (ALARA). (Section R1.1)
Proper foreign material exclusion control was demonstrated and good
radiological work practices were demonstrated during retrieval of the fuel
handling tool from the spent fuel pool. (Section U3.E2.2)
On August 5,1996, an unauthorized entry was made into the Millstone
Station protected area (PA) by an administrative contract person. The
individual had worked at the station, inside the PA until her previous
assignment ended on July 19,1996. Due to an oversight, she did not
surrender her badge and key card upon termination, although her key card had
been deactivated. When she arrived at the access control center, a co-worker
saw that she was having trouble entering through the access portal and used
her own valid key card and hand geometry to allow the individual to enter.
The co-worker followed the unauthorized individual into the PA by keying in a
second time. This issue is unresolved pending completion of the licensee's
corrective actions and further NRC review. (Section S1.1)
Review of the licensee vehicle barrier system, conducted in accordance with
NRC Inspection Manual Temporary Instruction 2515/132, " Malevolent Use of
Vehicles and Nuclear Power Plants," disclosed that the system was installed
and was being maintained in accordance with applicable regulatory guidance
and requirements. (Section S8.1)
On August 13,1996, representatives of NU, along with their attorney
conducted a drop-in meeting with the NRC staff, discussing their Unit 1
radwaste system investigation. NU's review uncovered no evidence that
suggests that any member of the NU staff intentionally communicated
inaccurate information to NRC inspectors. However, their investigation did
indicated problems associated with less than comprehensive answers to those
viii
- . _ - - . -
. . - . . - - . . . . . . . . . . . . - - . .. . . ....-- . _ - - - . . . .
. - . - - .
.
.
.
questions; inadequate management accountability; inadequate establishment
- of management expectations regarding communications with the NRC and
regarding facility conditions; and inadequate closure of a PIR that discussed
radwaste conditions. (Section X3.1)
i
e
[
l
IX
.
_ -
. ,
.
,
.
i
Report Details
Summarv of Plant Status
j
Unit 1 remained in an extended outage for the duration of the inspection period. The
licensee continues to review the plant's level of compliance with regulatory requirements,
and compliance with their established design and licensing basis, associated with an NRC
request pursuant to 10 CFR 50.54(f) and Confirmatory Order.
U1.1 Operations
U1.01
Conduct of Operations
O 1.1
General Comments (71707)
Using Inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing
plant operations. In general, the conduct of operations was professional and safety-
conscious; specific events and noteworthy observations are detailed in the sections below.
01.2
Coanizance of Operability Determination
a.
Insoection Scope (71707)
A routine review of an operability determination (OD) was performed. The ODs are used to
assess degraded plant conditions which affect equipment operability and provide
compensatory measures as applicable.
b.
. Observations and Findinas
A copy of a recently issued operability determination (OD), on a potential seismic
interaction with control rod blades and spent fuel racks, was requested from the Shift
Manager. The Shift Manager had difficulty finding the OD and both the on-duty and the
on-coming Shift Managers appeared to be unaware of the disposition of the issue.
The inspector determined that neither Shift Manger had reviewed the OD and that in
practice ODs, some of which contain compensatory actions, were not being reviewed prior
to assuming on shift duties.
Following a discussion with the operations manager and the Unit Director, the Shift
Managers were provided the expectation that new ODs would be reviewed prior to
assuming watchstanding duties and all ODs with compensatory actions would be reviewed
during each shift turnover. Further, the shift turnover checklist was revised to include ODs
as a topic of review and discussion. An adverse condition report (ACR) was also initiated
to document and track the resolution of this issue. Additional enhancements to the
computer tracking tools for system status and procedures such as the " Conduct of
Operations" and " Operability Determinations" are planned. The inspectors verified that all
ODs with compensatory actions were being reviewed during subsequent shift turnovers.
-
_
-
,
.
.
2
c.
Conclusion
Licensed operators were not cognizant of operability determinations which assessed and
i
dispositioned degraded plant conditions. Further, no process or control existed to ensure
that licensed operators review and remain cognizant of ODs, including compensatory
actions. The failure of licensed personnel to maintain cognizance of degraded conditions
that affect operability could adversely impact the ability of the on-shift personnel to assess
subsequent equipment degradation.
01.3
Fire in Unit 1 Drvwell
l
a.
Inspection Scope (93702)
On June 8,1996, a small electrical fire occurred in the drywell as a result of a poor cable
coupling for a ground cable on a piece of welding machinery. The fire was extinguished
within minutes by the fire watch and a worker in the vicinity of the welding activity. The
fire was extinguished through the application of dry chemicals. The inspectors reviewed
this event, as well as the circumstances leading up to the event and recovery activities
following the fire.
b.
Observations and Findinas
On June 6,1996, the radiation monitoring systems that cause the isolation of normal
ventilation systems in the reactor building and initiate the standby gas treatment (SBGT)
system were declared inoperable. The radiation monitoring systems were found to be
inoperable due to discrepancies with the associated surveillance testing. Technical
Specifications (TS) required the calibration of these instruments to include a response time
verification, which had never been performed up to this time. With the monitors
inoperable, the limiting condition for operation (LCO), required isolation of the normal
ventilation systems and initiation of SBGT within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The normal ventilation was
secured and the "B" train of the SBGT was placed in service in accordance with TS
3.2.E.2. Later that day, the Shift Manager received a preliminary operability determination
(OD) documenting the operability of the radiation monitoring systems in the current plant
configuration. The LCO was then exited, securing SBGT and placing the normal ventilation
system in service.
On June 7,1996, the inspectors received a copy of " Operability Determination, ACR No.
M1960020," written to justify why the radiation monitors were operable. After reviewing
the OD, the inspectors discussed the validity of the basis for operability with the Unit
Director. The Unit Director agreed that the OD did not provide a valid basis for concluding
operability, and informed the inspectors that he would have the Shift Manager declare the
radiation monitors inoperable. At 2:30 pm, that same day, the radiation monitoring
systems were again declared inoperable, the normal ventilation system was isolated and
"B" SBGT was initiated to comply with the technical specification action statement.
-
__
. . -
.
_-
_
-.
_ - - _. - . -
_ - -.
-
- _ . _ - _ -
-
_ . -
,
'
.
i
3
'
On June 8,1996, a srnali electrical fire occurred in the drywell, which lasted several
minutes, and was extinguished with a dry chemical handheld fire extinguisher. The fire
was the result of a poor connection at a cable coupling for a ground cable on a welding
machine. Approximately four hours after the fire was extinguished, a plant equipment
operator noticed that the differential pressures (D/P) across the running SBGT train were
high. The control room operators surmised that the filters in the SBGT train were clogged
as a result of the residual dry chemical powder in the drywell atmosphere. The control
room operators determined that the D/Ps were unacceptable by comparing the readings to
the surveillance test requirements. After conferring with the Unit Director and the Duty
Officer, the Shift Manager elected to unisolate the reactor building, start normal ventilation
and secure the running train of SBGT. The drywell atmosphere was ventilated using the
normal ventilation system for approximately 25 minutes. The "A" train of SBGT was
started and normal ventilation systems were secured and isolated. These actions were
taken to remove or " purge" the remaining dry chemical powder in the atmosphere via the
unfiltered normal ventilation system and thus prevent fouling of the redundant SBGT filter
train. However, this action resulted in a deviation tiom the requirements in technical
4
specification 3.2.E.2.
'
c.
Conclusions
This event is currently under review by the NRC staff. This issue is unresolved (URI
245/96-06-01) pending further NRC review and inspection.
U1.08
Miscellaneous Operations issues (92700)
08.1
(Closed) LER 50-245/96-01: performing work with the potential of draining the
reactor vessel during fuel movement. This event was discussed in Inspection
Report 50-245/95-42. No new issues were revealed by the LER.
U1.Il Maintenance
U1.M1
Conduct of Maintenance
M 1.1
Scent fuel Pool Tri-Nuclea.- Filter Removal
a.
Insoection Scope (62703)
On July 16,1996, plant personnel were removing a temporary filter assembly from the
spent fuel pool when a wire rope, attached to the filter assembly, was entangled with
control rods that were suspended from the spent fuel pool equipment rail. This caused the
control rods to shift position away from the wall and come to rest against an adjacent
spent fuel rack. The inspector proceeded to the refuel floor to evaluate the significance of
the event and to observe actions taken by the licensee to stabilize the temporary filter and
control rods. The inspectors reviewed this event, conducted interviews, inspected
associated documentation for the work activities, and evaluated the licensee's root cause
analysis for the event.
. _ _
.
4
"
l
4
'
b.
Observations and Findinas
On July 16,1996, maintenance and health physics personnel were attempting to remove a
portable "Tri-nuc" filter assembly from the spent fuel pool floor when a 1/8 inch wire rope,
attached to the filter assembly, caught on the bottom of three used control rod blades that
were stored along the east wall of the spent fuel pool. This caused the control rods to
move, resulting in a total of five control rods shifting position, moving away from the wall
j
clustering together, and coming to rest against an adjacent spent fuel storage rack. Upon
'
noticing the control rod movement, maintenance personnel stopped the crane. The "Tri-
nuc" filter was suspended at a point were the top of the assembly just broke the surface of
the spent fuel pool. The control rods were attached to an equipment rail at the top of the
spent fut:1 pool, each suspended by a cable, and resting perpendicular on the floor of the
fuel pool. One of the rods that was caught on the wire was lifted three feet off the fuel
i
pool flocr with one end resting close to the fuel pool liner and the other against an adjacent
spent fuel rack. The elevation of this rod caused its hook to become disengage at the top
of the control rod. The workers made an attempt to reconnect it, but were not successful.
After soveral attempts to untangle the control rods and free the "Tri-nuc" were also
unsuccessful, the work was stcpped and the workers left the refuel floor and informed
plant management.
Six of the eight individuals involved in this evolution were con;aminated as a result the
filter removal event. They were successfully decontaminated and whole body counts
indicated no internal dose was received. During the event, no area radiation monitor
alarms were received and no airborne radiation was detected (see section R1.1).
The licensae placed additional rigging on the "Tri-nuc" filter and a cable was placed on the
control rod that was no longer attached to its normal suspended cable. The licensee
inspected the tangled group of control rods with under water cameras on July 18,1996,
to assess any potential damage to the fuel pool liner, or adjacent spent fuel racks. No
damage was identified at that time. Following the under water inspection, the suspended
control rods were stabilized with additional cables attached to the bottom of the rods and
secured to the refueling bridge. The licensee is currently developing a recovery plan.
I
c.
Conclusions
The NRC is continuing to review this event. At the end of this inspection period, the
licensee had not completed its review of the event, and therefore, long term corrective
actions have not been initiated. The recovery plan will be finalized once the long term
corrective actions are in place. This issue is unresolved (URI 245/96-06-02) pending
further NRC review and inspection.
M1.2
ISI Proaram Review
a.
Insoection Scoce
This inspection was to review and verify the licensees commitment to Generic Letter 88-01
for the augmented, ultrasonic (UT) inspection program.
-
1
1
5
b.
Observations and Findinas
The inspector reviewed the licensees response and commitments to Generic Letter (GL) 88-
01 "NRC Position on IGSCC in BWR Austenitic Stainless Steel Piping." GL 88-01 was
issued by the NRC on 1/25/88, to assure that licensees inspection programs conform with
Title 10 of the Code of Federal Regulations (CFR) Part 50, Appendix A, General Design
Criterion (GDC) 4,14, 30, 31, and 32.
The inspector selected a sample of component records to verify component categonzation,
which is based on guidance in GL 88-01 and NUREG- 0313 " Technical Report on Material
'
Selection and Processing Guidelines for BWR Coolant Pressure Boundary Piping." The
components were categorized by the licensee based on the susceptibility to intergranular
stress corrosion cracking (IGSCC). Each category has a different examination schedule.
Category A components receive the least frequent examination, 25% of the components
examined once every 10 years; Category D components must be inspected every other
refueling cycle. Chiegory A components are made from corrosion resistant material and
are considered not susceptible to IGSCC. Category D components are considered
susceptible to IGSCC. The components selected for review were appropriately
categorized, according to GL-88-01 and NUREG-0313.
The inspector reviewed the licensees program for performing augmented UT examinations
for IGSCC. Three prograramatic weakness were identified. First, the IGSCC program does
not specify a methodology to evaluate unresolved UT indications (UIR's), in at least 16
cases, the methodology used to evaluate the UIR's was inappropriate. The indications
previously re-evaluated and overturned by the Level Ill, were later determined to be actual
flaws. The examination procedure and calibration blocks for the UT examinations are not
specified. Specifying the correct procedure and calibration block may prevent incorrect
guidance to perform the examinations. Finally, the IGSCC program does not provide
guidelines for tracking and trending UT indications. IGSCC indications were detected in
twelve pipe welds during previous inspections, as early as 1984, and not tracked for
subsequent inspection. In one case, weld RCBJ-7, four (4) IGSCC indications were noted
during the 1987 refueling outage (RFO). In the 1989 RFO, the welds were reported to
have no recordable indications (NRI). In 1995, RFO 15, two of the four circumferential
indication reported in 1987, were identified and rejected per the American Society of
Mechanical Engineers (ASME) Code,Section XI.
One unresolved item in report 50-245/96-01 was that the NDE Level!!I technician
ovorturned the UT Level 11 evaluation of IGSCC indications and returned the components to
service without an engineering evaluation. The NRC inspector reviewed 12 UIR's from
RFO 15 in which the Level ll evaluated the indications to be flaws and the Level ill re-
evaluated the indications as not being flaws, but as being geometrical reflectors. In these
cases, the evaluations of the UlR's were appropriate, and were dispositioned using
reasonable methods.
During the RFO 15 inspection, thirty five (35) welds were examined containing IGSCC.
Fourteen of the 35 welds were previously identified by the UT Level 11 technician, as early
as 1984, as having at least one IGSCC indication, and were subsequently overturned by
.
.
6
the NDE Levelill. The disposition of the 35 welds with IGSCC, during RFO 15, was to
replace the component or weld overlay the components prior to start-up.
Six reactor coolant components, (RCAJ-2, RCBJ-1 A, RRJJ-4, RREJ-4, RRCJ-4 and CUBJ-
18) with flaws were placed inservice, between 1984 and 1995, without flaw analysis arc
required b/ ASME Section XI,1986 Edition, Paragraph IWB-3640. The six components
were ultrasonically (UT) inspected between 1984 and 1994. During the UT examinations,
each component had at least one IGSCC indication. The ASME Section XI analysis was
not performed on the components because the UT Level Illinappropriately evaluated the
IGSCC indications to be geometry The indications were determined to be cracks during
refueling outage 15. The licensee performed an evaluation during the November 1995
refueling outage, RFO 15, in accordance with ASME Section XI,1986 Edition, IWB-3640,
to determine the operability of the components. The licensee determined the components
did not meet the requirements for continued service and declared the components
inoperable. Th3 licens
Mfined inoperability of a component as a decrease or elimination
j
of the operating safety mmgin for structural integrity. The licensee determined the safety
margin is decreased when a crack through wall dimension in the component is equal to or
greater than 75% of the pipe wall nominal thickness.
The six components had intergranular stress corrosion cracks (IGSCC) that were greater
than 75% through wall. Two of the six components leaked during preparation for weld
overlay. The reactor coolant systems were degraded to the extent a detailed evaluation
was necessary to determine system operability. The results of the licensees evaluation
determined the six components had an unacceptable structural integrity and a high
probability of abnormal leakage.
Five Category D (susceptible to IGSCCI welds (ICBC-F-18, ICBC-F-16, ICBC-F-14, RCAJ-6,
and RC9J-12) were selected for independent ultrasonic examination by the NRC mspector.
The welds were recently examined and accepted by NU. NU equipment and UT
procedures, MP-UT-2 and NU-LW-1, were used to examine the welds. The results of the
licensees examinations closely matched those of the NRC examinations, within the
expected to!erances. The NRC inspector detected a UT indication in weld ICBC-F-16 which
was not recorded by the licensee. The indication was 1.5" long, approximately 0.10" in
depth at 22" from top dead center. The indication information was turned over to the
licensee by the NRC for resolution. The licensee investigated the indication with
automated UT, manual UT, and radiography. The licensees concluded the indication was
caused by inclusions (stringers) in the base material,
c.
Conclusiorg
The components reviewed during the current outage were appropriately categorized for
IGSCC susceptibility and examination. The components were categorized based on the
guidance provided in GL-88-01 and NUREG-0313. Documentation detailing component
categorization was readily available for review by the inspector.
T.aee vweakness were identified during the program review. The program lacks detail to
prevent inadvertent procedure oversights during the examinations and evaluations. These
weakness resulted in UT indications being incorrectly overturned, and indications detected
<
,
l
l
.
7
in one inspection, being missed in a subsequent inspection. The method used to evaluate
UIR's relies solely on the NDE Level lil's expertise. These weakness could result in flawed
components being returned to service without engineering evaluation.
l
l
The UIR evaluation and resolution for RFO 15 was found to be appropriate.
The NRC inspector performed five independent UT examinations selected from Category D
components. With the exception of weld number ICBC-F-16, the results of the licensees
examination closely matched those of the NRC; within the expected tolerances.
It was determined during the November 1995 refueling outage, RFO 15, six inoperable
reactor coolant components, (RCAJ-2, RCBJ-1 A, RRJJ-4, RREJ-4, RRCJ-4 and CUBJ-18)
were previously placed inservice with unacceptable structural integrity and a high
probability of abnormal leakage. The six components had intergranular stress corrosion
cracks (IGSCC) that were greater than 75% through wall. Two of the six components
,
leaked during preparation for weld overlay. These flawed welds and the associated
weaknesses discussed above are considered unresolved pending licensee corrective actions
and further NRC review. (URI 245/96-06-03)
M1.3
Review of Uodated Final Safety Analvsis Reoort (UFSAR) Commitments
A recent discovery of a licensee operating their facility in a manner contrary to the Updated
Final Safety Analysis Report (UFSAR) description highlighted the need for a special focused
review that compares plant practices, procedures and/or parameters to the UFSAR
description.
While performing the inspection discussed in this report, the NRC inspector reviewed the
applicable portions of the UFSAR, Section 5.2.4, that related to the areas inspected. The
inspector verified that the UFSAR wording was consistent with the observed plant
practices, procedures and/or parataeters.
'
U1.Ill Enaineerina
U1.E8
Miscellaneous Engineering issues
E8.1
(Closed) Unresolved item 50-245/96-05-07: Nonconformance Reports
a.
Inspection Scone (37551)
A review of the implementation of the licensee's NCR process was performed as it related
i
to control of equipment operability. It appeared that the licensee had used the NCR
process, exclusively in some cases, to identify conditions adverse to quality on installed
plant equipment (i.e. in field deficiencies) contrary to procedure 3.05, Nonconformance
Reports. The NCR process did not require a prompt assessment of operability for these
degraded or nonconforming conditions on installed plant equipment. At the end of the
previous inspection period, all NCRs had not been dispositioned and assessed for
operability. URI 245/96-05-07 was established to review the NCR process,
.
-
i
8
b.
Observations and Findinas
The licensee conducted a review of open NCRs assess their impact on system operability.
At the time of the review, there was a backlog of 334 open NCRs. Open status was
defined as any NCR that was in the review or dispositioned status (i.e. not completed and
returned to Ouality and Assessment Services (QAS)). The licensee performed an initialline
by line review of the data base NCR descriptions to determine if they could adversely
affect operability either presently or in the past. This initial review identified 118 NCRs
that would require further assersment to determine if there was an operability concern.
The licensee identified, early in their review, that they could not easily determine the status
of these 118 NCRs since some of the original NCRs could not '
-hysically located.
Nuclear Group Procedure (NGP) 3.05, Nonconformance Report , states that QAS maintains
the NCR log which is a hand written log that lists the sequential number, originator, and a
brief description of the NCR. Once a number is assigned to an NCR, a copy of the form is
made and filed by QAS. A second computerized data base called the Master Tracking
,
Form (MTF), also maintained by QAS, but not addressed by the procedure, is used to track
i
the NCRs to completion.
From late June 1996, until mid July 1996, a search was conducted to account for all 118
NCRs. Of the 118 NCRs reviewed, no NCRs were identified as currently impacting system
operability. However, during the review, three NCR numbers were identified as gaps in the
accounting system; they did not appear on the open or closed list. There was no physical
paperwork associated with these three NCR numbers, nor was there a description or
originator listed in the log or MTF data base. Additionally, two adverse condition reports
(ACRs) were written during the licensee's review of the NCR process. ACR M1-96-0149
documented a situation where original NCRs were appearing in Work Planning for closure
after copies of the NCRs had been previously used to close the same NCRs. Further,
investigation by the licensee indicated that the disposition of the original and the copy
were different for three of the NCRs. ACR M1-96-0198 documented the fact that an
inspection of the open/ closed MTF NCR data base when compared with the NCR log,
indicated that the MTF data base did not accurately reflect all NCRs.
The inspector performed an independent review of the 118 NCRs for operability impact.
While no NCRs were identified as impacting system operability, the inspector noted seven
NCRs that were dispositioned, during this review, as having no current impact on
operability but would require repair / resolution prior to declaring the systems operable.
Since ACRs were not written to account for these NCRs, there is no process controls to
ensure all degraded and nonconforming conditions are tracked and corrected prior to
restoring systems to an operable status. This weakness was identified in NRC inspection
report 96-05, and is currently under review by the licensee. Additionally, the inspector's
review confirmed the licensee's practice of using the NCR process to identify conditions
adverse to quality on installed plant equipment (i.e. in field deficiencies) contrary to
procedure 3.05. NCRs were used to identify such deficiencies as a degraded concrete
base beneath a service water pipe support; degraded and leaking resin transfer piping;
circulation pump discharge headliner pitting due to corrosion; and wood block shims in the
hypochlorite system. NGP 3.05, Nonconformance Reports, section 6.1.1, states that "in
the field, the NCR is not used to identify deficiencies but to provide engineering direction to
the field when a condition adverse to quality cannot be made to conform to requirements
. - . . - _ _ _ .
. _ . _
_ . _ _ _ _ . _ -
.
-__
_ __
.
__ _ ___._._ .
.
!
l
.
'
9
l
or when an organization requires engineering direction concerning an identified deficiency.
In the field deficiencies are identified by trouble reports, automated work orders, ACRs,
j
surveillance, inspections, or audits." The vulnerability associated with the mis-application
of the NCR process is that it circumvents other plant processes, which would provide the
i
controls necessary for prompt operability determinations and to ensure all degraded and
'
!
nonconforming conditions are tracked and corrected prior to restoring systems to an
'
l
operable status.
j
i
c.
Conclusion
j
i
The NCR process was used as an identification process for degraded and nonconforming
conditions in field contrary to procedure NGP 3.05.10CFR 50, Appendix B. Criterion XV,
Nonconforming Materials, Parts, or Components which do not conform to requirements in
,
order to prevent their inadvertent use or installation. This is an apparent violation of
g
10 CFR 50, Appendix 8, Criterion XV. (eel 245-96-04)
l
Additionally, a review of a selected group of NCRs, for operability determinations, indicated
i
that the physical control of the NCR process was lost at Millstone Unit 1. The licensee
j
plans to review the remaining open NCRs (approximately 200) to account for each NCR,
j
either with originals or copies. At the end of this inspection report period, this had not
i
been completed. Unresolved item 245/96-05-07 is closed, and will be tracked with the
corrective actions associated with the foregoing violation.
4
s
4
.
.
.
10
Report Details
Summarv of Unit 2 Status
i
Unit 2 remained in cold shutdown throughout the inspection period. The unit has been
'
shut down since February 20,1996, due to uncertainty with the licensee's compliance
with the plant design and licensing bases. A comprehensive recertification process is
being conducted to support plant restart.
U2.1 Operations
U2.01
Conduct of Operations
01.1
General Comments (71707)
Using Inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing
i
plant operations. In general, the licensee's conduct of operatiens was professiorial and
"
safety-conscious; specific events and noteworthy observations are detailed in the sections
below.
U2.03
Operations Procedures and Documentation
1
03.1
Reactivity Controls Durina Plant Cooldown
a.
Insoection Scope
The inspector noted that on July 7,1996, operators injected boron into the reactor coolant
system (RCS) from the boric acid storage tank (BAST) without first sampling the tank to
verify boric acid concentration. The inspector was concerned that changing core reactivity
using an unverified boron source could result in an inadvertent RCS dilution,
b.
Observations and Findinas
,
While on shutdown cooling, the licensee injected boric acid from a BAST to raise
concentration in the circulating portion of the RCS in preparation for a core offload. Only
one of the two BASTS is used for this evolution because the other BAST satisfies the
4
technical specification requirement as an emergency boration source. In situations where
borating the RCS will require more than the volume of one BAST, it has been the licensee's
practice to make batch additions of boron to the BAST while injecting from that tank.
Although the initial BAST volume is recirculated and sampled, the BAST concentration is
considered unverified following the first batch tank addition. Discussions with operators
indicated that batch additions are performed without sampling due to time constraints,
particularly during a plant cooldown. Since the RCS is not borated to the cold shutdown
concentration prior to commencing the cooldown, boric acid must be added at a sufficient
rate to ensure adequate shutdown margin is maintained. The operators contend that
adequate shutdown margin can be verified through frequent RCS sampling. They also
contend that cooling down in parallel with borating minimizes radioactive waste because
the boric acid is injected as coolant volume contracts. In addition, operators stated that
_.
_
.
.
11
there are sufficient controls in place to prevent dilution due to the fact that the. boric acid
powder is packaged in QA-inspected, pre-measured bags.
The inspector had several concerns with the licensee's RCS boration practices including:
,
(1) The licensee statec' that the batching methods in procedure OP 2304C, "Make Up
(Boration and Dilution) Portion of the Chemical and Volume Control System," are designed
,
to maintain BAST concentration between 2.5 percent and 3.5 percent by weight boric
acid. This equates to 4371 ppm to 6119 ppm, which is a very wide concentration band.
This greatly limits operator control of core reactivity changes, in addition, it allows for an
inadvertent dilution due to human error; (2) The Final Safety Analysis Report (FSAR),
Section 9.2.3.3, states that "the boron concent;ation is increased to the cold shutdown
value prior to the cooldown of the plant. TNs is done to assure that the reactor has an
adequate shutdown margin throughout the cooldown." However, procedure OP 2207,
,
" Plant Cooldown," Step 4.1.3, states that "it may not be possible in all situations to borate
to cold shutdown concentration before commencing cooldown." It is the licensee's normal
practice to borate to cold shut concentration concurrently with a plant cooldown, and; (3)
FSAR, Section 9.2.3.3, states that "the operator does not insert the shutdown group of
[ control element assemblies] CEA's until the cooldown is completed and until he verifies
the concentration of boron in the reactor coolant by sample analysis." However, procedure
,
OP 2206, " Reactor Shutdown," Step 4.3.3, inserts all control rods in the shutdown group.
Procedure OP 2207, Section 2, " Prerequisites," step 2.1.1, specifies that the reactor is
shutdown with all control rods fully inserted. The FSAR is also not consistent with
Technical Specification 3.1.3.7 which states that the control rod drive mechanisms shall be
de-energized in modes 3,4,5 and 6 whenever the RCS boron concentration is less than
,
the refueling concentration.
'
1
c.
Conclusion
10 CFR 50.59, " Changes, Tests and Experiments," states that a licensee may make
changes in the facility and procedures as described in the safety analysis report without
prior Commission approval, unless the proposed change involves a change in the technical
specifications or an unreviewerl safety question. Records for these changes must include a
written safety evaluation, which provides the basis for the determination that the change
does not involve an unreviewed safety question. FSAR Section 9.2.3.3 requires that
boron concentration be increased to the cold shutdown value prior to cooldown and that
the shutdown group of control rods remain withdrawn until the cooldown is completed and
boric acid concentration verified. The failure to prepare a safety evaluation to reflect the
changes to the facility implemented in procedures OP 2206 and OP 2207 is an apparent
violation. (eel 336/96-06-05) These concerns are safety significant in that they constitute
a degradation of barriers to the prevention of an inadvertent criticality.
Additionally, control and monitoring of reactivity changes has been the subject of an
ongoing licensee and NRC concern. This issue was included in the " improving Station
Performance" Plan to assure sitewide, comprehensive corrective action. The' licensee's
practice of injecting from the BAST before sampling is not consistent with positive control
practices with reactivity changes. This practice should be addressed when responding to
the apparent violation.
- .
-- _
- - - .
-
.
_-__
. _ . - -
. . .
__
.. ~.
_ - - _ .
.
.
I
12
U2.07
Quality Assurance in Operations
07.1
Proaram for Submi'.tal of Licensee Event Reports
During this inspection, the licensee's program for review and submittal of licensee event
I
reports (LERs) was evaluated by assessing 19 of the LERs submitted for Unit 2 in 1995
and 1996. The reports, in general, were timely and informative, but many required
supplemental information because either causal analysis had not been complete or
corrective action plans were not finalized. The inspector determined that 3 LERs (95-19,
96-01, and 96-03) contained commitments that were not completed on schedule; and 4
LERs (9519,95-41,96-01, and 96-12) did provide adequate corrective actions to address
the event causal factors. These deficiencies are detailed in other sections of this report.
The inspector considered the licensee's program for development, review and tracking of
LER issues to be weak because these discrepancies were not routinely identified and
corrected by the licensee,
U2.08
Miscellaneous Operations issues (92700)
l
08.1
Update URI 336/95-42-03: Closed LER 50-336/96-01: Reactor Coolant
l
System (RCS) Heatuo Rate Exceeded Technical Specification Limit
This event was documented in NRC Inspection Report 50-336/95-44. The licensee
l
identified this occurrence during an event review team (ERT) review of several other
incidents involving exceeding RCS heatup and cooldown rates. An engineering evaluation
concluded that no adverse consequences to the reactor coolant system occurred as a
result of this transient. The LER was supplemented on June 27,1996 vice April 2,1996,
as committed. The inspector found the LER to be incomplete because it did not discuse,
why the plant monitoring program for cooldown and heatup rates was inadequate, and
,
what corrective action was implemented to resolve this causal factor. The violations of TS
requirements will be evaluated upon completion of the ERT corrective actions and licensee
update of the LER. This issue is tracked by unresolved item URI 336/95-42-03.
!
l
08.2
Closed LER 50-336/96-02. 03. 04 and 05: Service Water Strainer Backwash
l
System inadeauacies
i
A common discharge pipe from all the strainer backwash valves froze during cold weather.
l
This rendered both service water systems technically inoperable for 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />. These LERs
l
describe a series of deficiencies in system design and operator performance. NRC
Inspection Report 50-336/95-44 documents review of the service water strainer
l
inadequacies. Licensee corrective actions were not fully effective requiring subsequent
enhancement. Two errors noted in LER 96-02 were subsequently corrected in LERs 96-03
and 96-04. Two corrective actions in LER 96-03, (to implement an operability
determination procedure and supplement the LER upon completion of the event review
team) were not completed as committed. The NRC is tracking completion of corrective
actions and disposition of the potential enforcement actions as eel 336/95-44-05.
1
- . - ..
.
__
. . -
-
-
. _ -
-
4
"
.
13
08.3
(Closed) LER 50-336/96-07: Reactor Coolant System (RCS) Cooldown Rate
!
Exceeded the Technical Specification Limit
This event was documented in NRC Inspection Report 50-336/96-01. The licensee
identified several prior instances of RCS cooldown rate exceedance during an event review
1
team (ERT) investigation of incidents involving RCS heatup rate exceedance. An
'
engineering evaluation concluded there were no adverse consequences to the cooldown
rate exceedance. These prior violations of TS requirements will be evaluated upon
completion of the ERT corrective actions. This issue is tracked by unresolved item URI
i
i
336/95-42-03.
u
!
U2.ll Maintenance
j
U2.M1
Conduct of Maintenance
i
j
M1.1
General Comments (62703, 61726)
!
Using Inspection Procedures 62703 and 61726, the inspectors conducted frequent reviews
of ongoing plant maintenance. In general, the conduct of maintenance and surveillance
f
activities was professional and safety-conscious; specific events and noteworthy
1
i
observations are detailed in the sections below.
l
U2.M8
Miscellaneous Maintenance issues
.
i
M8.1
(Closed) LER 50-336/95-40: Late Surveillance of the Reactor Protection
j
System
On October 22,1995, the licensee found that the technical specification daily surveillance
,
j
SP 2601D, " Power Range Safety Channel and Delta T Power Channel Calibration," had not
been performed within its required frequency. The surveillance was performed
satisfactorily later that day. The late performance of surveillances is discussed in detailin
,
NRC Inspection Reports 50-336/95-38 and 50-336/96-04 and is being tracked by violation
i
'
336/95-38-01.
M8.2
(Closed) LER 50-336/95-41: Potential for Emeraency Diesel Generator
i
Overload Durina Surveillance
l
a.
Inspection Scooe
.
i
The inspector reviewed the licensee's response to LER 50-336/95-41 and evaluated
I
whether their corrective actions satisfied inservice Test Program requirements.
!
<
}
b.
Observations and Findinas
Prior to performing procedure SP 21136, " Safety injection and Cantainment Spray System
Valves Operational Readiness Test," an operator noted the potential to overload an
emergency diesel generator (EDG) if a loss of coolant accident (LOCA) and a loss of normal
'
i
power (LNP) event were to occur while the surveillance was in progress. To perform
-
. .
. - -.
.
. .
-
-.
- . _ - - -
.
_ - - .
. . - . _ .-
.-
.
.
N
14
,
4
backflow testing of the high pressure safety injection (HPSI) pump discharge check valves,
the swing HPSI pump is mechanically and electrically aligned alternately to each facility
'
(equipment, devices, cables and raceways have an assigned number that indicates if they
are in vital service or not. These numbers are called the " Facility Codes"). This places
two HPSI pumps on one facility, rather than the one pump assumed in the EDG loading
analysis. If two HPSI pumps were to start at EDG loading sequence 1, generator voltage
'
would dip to 72 percent of rated voltage which is lower than the 75 percent design limit.
In addition, the 3250 Kw 300-hour rated load would be exceeded at sequence 3 (3384
.
Kw) and sequence 4 (3604 Kw). The amount of time that two HPSI pumps are aligned to
a single facility is approximately two hours for each facility.
J
Millstone Unit 2 Second Ten Year Inservice Test Program contains an NRC approved Relief
'
Request No. IWV-6 which allows full stroke testing of the HPSI check valves when the
'
reactor vessel head is removed, because this is the only plant condition where full design
flow can be attained. The relief request states that partial stroke exercise of each HPSI
check valve will be performed each month. As a corrective action to address the EDG
loading concerns, procedure SP 21136 was changed to specify that backflow testing of
!
the HPSI pump discharge check valves would only be performed with the reactor vessel
head removed. Refueling Shutdown Justification (RFOJ-004) was prepared to change the
Inservice Test Program to address the change in HPSI check valve test frequency. The
justification states that "the valves will be partially stroked (open only) quarterly and full
'
i
stroke exercised (open and shut) during refueling outages when the reactor vessel head is
removed. NRC approved Relief Request IWV-6 has previously approved deferring full
stroke testing to refueling outages when the reactor vessel head is removed."
)
The licensee's premise that partially stroked means open only and full stroke exercised
means open and shut is inconsistent with the way the NRC and the licensee previously
j
defined these terms. The licensee's Inservice Test Program describes the NRC guidance as
'
stating that it is considered full stroke testing when the flow used is at least that which is
i
identified in the plant's safety analysis. Any less flow used will be considered as a partial
stroke unless it can be demonstrated that the lesser flow will still place the valve disk in
i
the same position as the flow in the plant's safety analysis. Therefore, Relief Request
2
IWV-6 addressed the fact that the check valves could only be tested fully open during
i
refueling outages, it did not allow the licensee to perform testing of the check valves in
the closed direction only during refueling outages. This is a concern because backflow
testing is important to ensure that the design HPSI flow is not diverted through idle pumps.
c.
Conclusion
Although NRC NUREG 1482, " Guidance for Inservice Testing at Nuclear Power Plants,"
allows licensees to make certain changes to their IST program that are consistent with
l
Code requirements, the licensee's basis for changing the back-flow test frequency of the
l
HPSI check valves is not supported by an inability to test the valve during operations. The
inspector was concerned that eliminating the quarterly backflow testing was unnecessary
3
because on-line testing can be performed without potentially overloading the EDGs; for
instance; by mechanically, but not electrically, aligning two HPSI pumps to one facility.
Such a relaxation is neither consistent with the Code nor covered by the approved relief
request. The licensee agreed to revise procedure SP 21136 to perform quarterly backflow
- _ _ _
_
.
.
15
testing. This issue is considered unresolved to allow the NRC to review the planned
changes to this procedure. (URI 336/96-06-06)
M8.3
(Closed) LER 50-336/95-42: Late Surveillance of the Off-Site Line Verification
On November 15,1995, the licensee found that the shiftly technical specification
surveillance to verify the alignment of two circuits between the off-site transmission
network and the switchyard had not been performed within its required frequency. The
'
surveillance was performed satisfactorily later that day. The late performance of
surveillances is discussed in detail in NRC Inspection Reports 50-336/95-38 and 50-
336/96-04 and is being tracked by violation 336/95-38-01.
'
M8.4
(Closed) LER 50-336/95-44: Late Surveillance of the Containment Personnel
Air Lock
On November 29,1995, the licensee noted that the 6-month surveillance for the
containment personnel air lock was not performed within its technical specification required
frequency. This LER was historical in nature and the surveillance has been satisfactorily
performed since this event. The late performance of surveillances is discussed in detail in
NRC Inspection Reports 50-336/95-38 and 50-336/96-04 and is being tracked by violation
336/95-38-01.
M8.5
(Closed) LER 50-336/95-45: Plant Shutdown due to Leakina Charaina System
Valves
On December 14,1995, Unit 2 was shut down to assess the structural integrity of valve
2-CH-435, a ncn-isolable valve which provides thermal relief for the charging side of the
regenerative heat exchanger. During a containment entry, the licensee found an active
body-to-bonnet steam leak on valve 2-CH-435 resulting in a boron buildup around each of
the four studs. The leak rate was estimated to be one drop per minute. Machinery history
showed that there was no past leakage and no prior maintenance on the valve. The unit
was shutdown to evaluate whether stud degradation had occurred as a result of the boric
acid buildup. The licensee found the condition of the studs, as well as the valve body,
valve bonnet, and seat rings, to be good. The root cause was an original installation error
by which the as-found torque on two of the four body-to-bonnet studs was less than the
manufacturer recommended torque.
Prior to returning the plant to operation, a sample of 20 similar valves was visually
inspected, and no signs of leakage or degradation were found. The licensee stated that
this provided assurance that no similar condition existed for other valves. The LER stated
that prior to startup from refueling outage RFO 13, the preload on similar valves will be
checked. The licensee is now in the process of completing the preload checks for these
valves during the current mid-cycle shutdown.
____ _______
_._
__ _ _ _ ..
- _ _ _ _ . _ _ _ _ _ _ _ _ _
.
i
4
i
i
16
4
i
i
M8.6
(Closed) LER 50-336/96-12: Valve Internals Missina Rendered Safety Function
,
,
Inocerable
l
This event involved a missing part in a solenoid valve that controls the position of safety
injection valve 2-SI-618. This LER did not address corrective actions for the test program
!
and work control causal factors of this event. A violation of test requirements was cited
i
for this event in NRC Inspection Report 50-336/96-04. The licensee committed to
,
supplement the LER by September 25,1996.
!
.
1
J
!
U2.lli Enaineerina
i
l
U2.E1
Conduct of Engineering
j
E1.1
Reactor Coolant System Boron Concentration to Supoort Core Off-Load
i
,
j
a.
Inspection Scoce
!
}
The inspector evaluated the licensee's plans to attain the required boron concentration in
1
the reactor coolant system (RCS) to support a full core off-load. During a normal
!.
shutdown and cooldown, if fuel movement is planned, the licensee borates the RCS to the
refueling concentration while the reactor coolant pumps (RCPs) are operating. This mixes
the boron throughout the RCS, and satisfies Technical Specification (TS) 3.9.1 which
requires that when the reactor vessel head is unbolted 'or removed, a uniform and sufficient
boron concentration shall be maintained in all filled portions of the RCS and the refueling
i
canal. Uniform and sufficient concentration (1730 ppm) is needed to ensure that the
i
4
reactor remains shutdown (without control rods) during the refueling process. Since the
4
licensee did not intend to off-load the core when the unit was shut down, the RCS was
j
boNted to only 1320 ppm to satisfy shutdown margin requirements for mode 5. The RCPs
i
j
were secured and shutdown cooling was initiated. After the shutdown, the licensee found
that the core must be off-loaded to effect repairs of a safety injection valve. Since the
-
shutoown cooling system takes suction from the #2 hot leg and injects into all four cold
l
legs, large portions of the hot and cold legs, as well as the steam generators and RCPs, are
.
not circulated (and thereby mixed) by the shutdown cooling system. Therefore, the
I
licensee evaluated methods to achieve the TS required uniform boron concentration in all
j
filled portions of the RCS.
l
b.
Observations and Findinas
l,
On June 3,1996, the licensee submitted to the NRC a proposed one-time revision to TS
3.9.1 that would strike the words "of all filled portions" and " uniform." In addition, a
j'
footnote was proposed stating that for this Cycle 13 mid-cycle outage, it was acceptable
j
for boron concentration in the steam generators and unmixed portions of the hot and cold
{
legs to initially be as low as 1300 ppm. The technical basis for the proposed change
i
concluded that if the shutdown cooling system was borated to greater than 1820 ppm, the
i
entire RCS would remain above the TS required concentration (1730 ppm), even if all the
I
water in the loops were at 1400 ppm boron and were mixed with the shutdown cooling
l
system water. The licensee would achieve the increase in boron concentration in the loops
j
from 1320 to greater than 1400 ppm by a partial drain and refill of the loops. To minimize
I.
.
n.w
. . , , . _
__.
_ . _ ,
_
,
_ , _ . . ,
.
.
17
any stagnant volume of RCS water at 1300 ppm, the licensee planned to refill the RCS
from the refueling water storage tank using the "B" charging pump which would pump
'
water through temporary hoses to an instrument tap on each of the four cold legs. Using
this method, it was possible that the subsequent samples would show that all filled
4
portions of the RCS were uniformly above the refuel boron concentration. Thus, the
potential existed that this evolution could eliminate the need for the TS change.
1
On June 19,1996, the inspector attended a plant operations review committee meeting
)
(PORC) in which a change to the Unit 2 Technical Requirements Manual was approved
which involved a clarification to TS 3.9.1 to better define the term " uniform." The licensee
'
planned to use this clarification to assess RCS samples that would be taken following the
draining and refilling evolution. The clarification recognized that some engineering
judgment is involved in determining whether the boron concentration is uniform because:
1) The TS provides no tolerance range for the term uniform; and,2) boron concentration
sample results have a margin of error of approximately 10 ppm. The TS clarification stated
that a " uniform" boron concentration is attained when all filled portions of the RCS and
refuel pool are greater than the required refueling boron concentration. The bases for the
clarification states that the " intent" of uniform in TS 3.9.1 is to ensure that the RCS and
water volumes having direct access to the reactor vessel and core are maintained greater
than the required refueling boron concentration.
The NRC noted that the licensee's clarification of the term uniform was too broad. The TS
clarification allowed boron at various RCS sample locations to differ greatly (for example
500 ppm or more) but would still satisfy the licensee's definition of " uniform" as long as
the samples were all above refueling boron concentration. The TS clarification was
unacceptable because samples with large differences in boron concentration could not
reasonably be considered uniform. Although a non-uniform RCS is not a safety concern as
long as all portions are maintained above the refueling boron concentration, the TS
requirements regarding uniformity must be satisfied until such time as a TS change is
approved. The licensee did not implement the approved TS interpretation.
The inspector also evaluated the licensee decision to drain the RCS to mid-loop. This
activity is considered one of the highest shutdown risk evolutions due to an increased
possibility of losing shutdown cooling, a shorter time to boi!, and a reduced water volume
above the core. Although there is currently no regulation that requires the licensee to
avoid mid-loop operation, the licensee should have a sound justification why using mid-loop
operation is the best option available, especially considering the tact that the "B"
emergency diesel generator was inoperable and the shutdown cooling system was
degraded by the stuck-open loop isolation valve. The licensee stated that the primary
reason for draining to mid-loop was to attempt to satisfy the existing TS requirements, and
thus e minate the time needed for the TS amendment process.
u
Following discussions with the NRC, on July 3,1996, the licensee submitted to the NRC
another revision to TS 3.1.9 tha proposed an alternative plan. The licensee proposed to
raise SDC volume boron concentiation to greater than 1950 ppm, and borate the reactor
vessel head area by lowering levei in tl e vessel to 1 to 3 feet below the flange (above the
defined reduced inventory precaution conditions), and refilling with water greater than
1850 ppm boron. This method did not involve mid-loop operation, and eliminated the need
.
.
_.
. _ .
__
-
.
18
for temporary hoses to circulate water into idle portions of the RCS loops. While awaiting
NRC approval of the proposed TS change, the licensee borated the SDC volume to greater
than 2 l00 ppm. Various RCS sample results indicated substantial diffusion or mixing of
boron into the idle loops. However, strict uniformity of the samples was not achieved. On
August 13,1996, the NRC approved a one-time TS change to allow entry into the
refueling mode without uniform boron concentration. At the end of the inspection period,
the licensee was finalizing plans for the full core offload.
c.
Conclusion
The fact that the licensee did not establish a uniform refueling boron concentration in the
RCS prior to securing RCPs was reasonable, because they could not have anticipated the
need to perform a core off-load. However, the licensee safety perspective in dispositioning
this problem was not conservative in that they planned to drain the RCS to mid-loop when
other options involving less risk were available. In addition, PORC did not provide rigorous
oversight in approving a TS clarification that redefined " uniform" boron concentration, such
that, while meeting the intent of the TS, it provided so much latitude that it would not
have complied with the TS as written.
E1.2
Refuelina Pool Drain Line
j
a.
Insoection Scope
On July 11,1996, the licensee prepared an adverse condition report (ACR) that addressed
the fact that non-seismic piping was connected to the refueling pool drain header that was
not isolated from the header during refueling. The inspector evaluated the licensee's
response to the ACR to determine if it was consistent with their response to NRC Bulletin
84-03, " Refueling Cavity Water Seal."
b.
Observations and Findinas
There are two 4-inch refueling pool drain lines each containing a manual isolation valve (2-
RW-123 & 124). The two drain lines join to form a common 4-inch header that directs
water outside containment to the suction of the refueling water purification pumps. The
refueling water purification pumps can discharge through an ion exchanger back to the
refueling pool or they can discharge to the refueling water storage tank to drain the
refueling pool. Between valves 2-RW-123 & 124 and the containment penetration, there is
31 feet of piping that is seismic Class ll (non-seismic).
Procedure OP 2305, " Spent Fuel Pool Cooling and Purification System," states that the
purification system should be operated continuously during refueling and accordingly,
specifies opening valves 2-RW-123 & 124. The ACR addressed the fact procedure AOP
2578, " Loss of Refuel Pool and Spent Fuel Pool Level," does not specifically state that
valves 2-RW-123 & 124 should be verified closed if a decrease in refueling pool level is
observed. Instead, procedure AOP 2578, states that "jf conditions allow, verify that a
cavity drain line has not failed." Another concern was that access was difficult because
the valves were 13 feet above the floor. As corrective actions, the licensee planned to
change procedure AOP 2578 to specify closing valves 2-RW-123 & 124 in the event of an
--
_.
-
--
. . - - . - .
_
_
. . -
__-
-
.
.-
-
>
.
<
19
1
l
unexpected decrease in refueling pool level. They also planned to specify the location of
I
the valves and how to access them.
l
The inspector was concerned that operator actions were being used to compensate for the
'
1
fact that a portion of the drain line was not seismically qualified. in addition, the licensee's
response to NRC Bulletin 84-03 was based on the premise that valves 2-RW-123 & 124
'
would remain closed thereby eliminating the need for operator actions. This was confirmed
by the engineer who prepared the design modification that upgraded the drain line piping
,
l
from the refueling pool to valves 2-RW-123 & 124 to seismic class 1. This explained the
lack of specificity in procedure AOP 2578 regarding the need to close the isolation valves.
-
.
.
1
The failure to specify that valves 2-RW-123 & 124 would remain closed during refueling
operations is a concern because: (1) Following a seismic event, maximum flow through a
broken drain line would be approximately 15001 ;m. The licensee had no evaluation to
demonstrate that operators would have sufficie
time to close the valves or whether
operators could even reach the valves with a no <by 1500 gpm leak; (2) Although the flow
rate through a broken drain line is less than the 6490 gpm flow rate associated with a
-
reactor cavity seal failure, the consequences are significantly worse because a cavity seal
j
failure would drain the refueling pool to the reactor vessel flange while a drain line failure
would also drain the south saddle and transfer canal. In addition, a drein line break would
'
also release a larger volume of water to the containment than is assumed in the loss of
coolant accident analysis which could result in the submergence of essential equipment.
'
l
The licensee had no evaluation that addressed this larger volume of water and no
j
procedure describing operator actions to be taken.
More importantly, drainage of the south saddle and transfer canal eliminates these areas as
safe fuel storage locations. With a cavity seal failure, three fuel assemblies could be safely
i
stored in the south saddle area of the reactor cavity (two in the upender and one on the
4
refueling machine in its full down position.) However, drainage of the south saddle and
transfer canal following a drain line break would expose these fuel assemblies which would
l
greatly increase radiation levels and possibly result in fuel damage. Procedure AOP 2578,
j
Step 3.6 states that a refueling cavity drain line failure in the south saddle would drain that
area completely and would eliminate this area and the transfer canal as a safe storage
-
location. However, Step 4.3.3, only verifies that a cavity drain line has not failed "if
l
conditions allow." This step is inadequate in that it does not reflect that this verification is
crucial in determining a safe fuel storage location.
Since the licensee's bulletin response was written with the presumption that valves 2-RW-
!
123 & 124 would remain closed, procedure AOP 2578 does not address the failure of non-
seismic refueling purification system piping and components outside containment. This
i
scenario is safety significant because it could result in reactor coolant leakage outside the
containment that would not be available for recirculation.
Since procedure AOP 2578 specifies that containment be evacuated once the fuel
l
assembly is lowered to a " safe" location, operators would not be inside containment during
i
the approximately 1 % hours that would be available to determine a drain line break had
occurred. The safe locations with a drain line break are the core and the transfer carriage
,
'
after it is moved to the spent fuel pool and the transfei tube isolation valve is closed.
,
Y
_ ,-
- -
. _
-
.-.
. _ -
. - - - - -
.
.
-. - .
.-.
.
f
f
.
20
Procedure AOP 2578 states that it takes approximately 50 minutes to completely close
this valve.
<
Although it is not mentioned in the licensee's bulletin response, the north saddle of the
refueling pool would also drain in the event of a drain line failure. Even if fuel assemblies
'
were safely stored, completely uncovering the reactor upper guide structure or lower
internal components that could be stored in the north or south saddle could significantly
increase radiation levels.
The inspector discussed the above concerns with licensee management who agreed to
danger tag closed valves 2-RW-123 & 124 during the upcoming refueling operations. The
licensee plans to utilize the submersible filtration unit that they normally use for refueling
pool purification and clarification.
c.
Conclusion
The refueling pool drain line iss .es that are discussed above are considered unresolved
pending further NRC review es the licensee's disposition of these concerns. (URI 336/96-
06 07)
U2.E2
Engineering Sapport of Facilities and Equipment
E2.1
Core Tilt Evaluation
a.
Insoection Scope (37550)
The inspector reviewed the core tilt technical specification surveillance tests for fuel cycles
12 and 13. The inspection focused on the actions taken in response to a tilt anomaly that
occurred at approximately 10,000 megawatt days per metric ton of uranium (MWD /MTU)
burnup during cycle 12 operation. The azimuthal power tilt is the maximum difference
between the power generated in any core quadrant and the average power of all
quadrants, divided by the average power of all quadrants of the core. The azimuthal power
tilt was determined by using the fixed incore flux detector system and the INPAX incore
i
analysis computer code,
b.
Observations and Findinas
A graph of the azimuthal power tilt as a function of core burnup was documented in
calculation C12-01181 F2, Rev. O, March 2,1995, " Millstone Unit 2 Cycle 12 incore Data
Analysis." The azimuthal power tilt was approximately .01 for fuel burnup ranging from 0-
10,000 MWD /MTU, with the largest tilt located in the upper half of the core. At a burnup
of approximately 10,000 MWD /MTU, the tilt began to slowly increase. The maximum
measured tilt was approximately .017, at 12,000 MWD /MTU. The azimuthal tilt then
slowly decreased to approximately .012 for the duration of cycle 12. The incore azimuthal
tilt angle moved from approximately the 75 degree angle to the 328 degree angle during
cycle 12. The rate of change of the azimuthal tilt angle began to accelerate at about
8,000 MWD /MTU. Technical specification 3.2.4 requires that the azimuthal power tilt not
__.
_ _ - _ _ .
_
_
,
.
.
21
exceed .02, in mode 1, at power levels greater than 50% power. The azimuthal power tilt
did not exceed the technical specification requirement during the cycle 12 transient.
l
An evaluation of potential causes for the increase in azimuthal power tilt was documented
in attachment 2 to calculation C12-01181-F2. An evaluation of the fixed incore detector
l
response data indicated that the tilt was not the result of instrumentation or calculational
l
errors. The evaluation concluded that fuel assembly RW-15 was the most likely cause for
I
the tilt. Fuel assembly RW-15 was a reconstituted fuel assembly that contained 5 stainless
l
steel pins in place of fuel rods. This conclusion was largely based on the observation that
the peak quadrant power rotated to the quadrant that contained fuel assembly RW-15.
This conclusion was not confirmed by a more detailed core analysis, which concluded that
fuel assembly RW-15 could not cause the magnitude of til observed.
A Plant Information Report (PIR) 2-94-059, " Azimuthal Power Tilt increasing" (dated
2/11/94), was written to initiate a root cause evaluation for the higher than expected
azimuthal power tilt. The root cause evaluation for the PIR was performed by Reactor
Engineering, Nuclear Fuels Engineering and the fuel vendor. Several potential causes were
evaluated including improper instrument operation, misloading of the burnable poison, and
a separated control rod finger. The investigation was unsuccessful in identifying a root
cause. The PIR was closed on July 18,1994, with no recommended corrective actions.
A full core computer design code was used to evaluate the potential causes for the tilt.
The steel pins in the reconstituted fuel assembly RW-15 and misloading of the burnable
poison (gadolinium) were both evaluated as potential causes for the tilt. The burnable
poisons were suspected because tha gadolinium poison was expected to burnout at
approximately the core burnup where the tilt increase occurred. The results of the analysis
(Memorandum, "MP-2 Cycle 12 Core Radial Power Tilt," dated February 10,1995) were
inconclusive. Both the steel pins and burnable poison misloading predicted a change in the
azimuthal tilt angle similar to that experienced during the tilt transient; however, the
magnitude of the tilt was much less than the measured tilt. The fuel vendor verified that
there was no misloading of burnable poisons by reviewing manufacturing records. The fuel
vendor was also unable to identify a root cause for the observed increase in tilt,
c.
Conclusions
The peak azimuthal tilt did not exceed technical specification limits during cycle 12 tilt
transient. The plants accident analysis is valid for azimuthal tilt values less than the
technical specification limit. The root cause for the increase in the tilt was not identified.
The initial determination that the reconstituted fuel assembly was the most likely cause of
the tilt was not substantiated by further analysis. The inspector concluded that the
credible causes for the increase in tilt were thoroughly evaluated and the depth of the root
cause analysis was appropriate.
o
.
22
E2.2
Estimated Critical Rod Position Calculations
a.
Insnection Scone (37550)
The inspector reviewed the cycle 12 and 13 estimated critical rod position (ECP)
calculations to ensure compliance with technical specification requirements. The ECP is a
reactivity balance used to estimate the boron concentration and control rod position where
criticality will be achieved during reactor startups. The actions implemented to improve the
accuracy of the ECP calculations were also reviewed.
b.
Observations and Findinas
The ECP reactivity calculations are performed in accordance with Operating Procedure OP-
2208, Rev.11, " Reactivity Calculations." Technical specification 4.1.1.1.2 requires that
the actual critical reactivity be within 1 % delta-k/k of the predicted value. Operating
Procedure OP-2208 conservatively requires that criticality be achieved within .9% delta-k/k
of the ECP prediction. The inspector reviewed the cycle 12 ECPs. In all cases the
technical specifications and the administrative limits of OP-2208 were satisfied. However,
the licensee was not satisfied with the magnitude of disagreement between the actual and
predicted ECP values and implemented actions to improve the accuracy of the ECPs.
A Plant incident Report (PIR) 2-93-101, dated May 25,1993, was written to document a
cycle 12 occurrence where criticality was not achieved prior to the control rods being fully
withdrawn. The immediate corrective actions were to validate the ECP calculation and the
input data. Following validation, the Reactor Engineering staff calculated a new ECP,
boron concentration was reduced, and the reactor criticality was achieved. The reactor
reached criticality approximately 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> following the original startup attempt. The
failure to establish criticality prior to the control rods being fully withdrawn was an
operational inconvenience; however, at no time during this startup were technical
specification 4.1.1.1.2 or associated administrative requirements exceeded.
In response to this PIR, the Nuclear Analysis Section provided several recommendations to
improve the ECP calculations and operating procedure (Memorandum, " Plant incident
Report 2-93-101," dated July 27,1993). The recommendations were to use revised
power defect curves and to ensure adequate control rod bite when calculating ECPs for
high xenon startups. Operating Procedure 2208 was revised to implement these
recommendations. The corrective actions were successful in preventing similar
occurrences during the remaining cycle 12 startups. A human error in performing an ECP
,
calculation was determined to be the cause of a cycle 13 occurrence where criticality was
'
not achieved prior to the control rods being fully withdrawn (Adverse Condition Report
04601).
The Nuclear Analysis Section staff continued efforts to improve the accuracy of the ECP
calculations. On April 28,1994, an improved ECP methodology (Calculation W2-517-405-
NA, Rev. 0) was submitted to Nuclear Fuels Engineering (NFE) management for review and
approval. The calculation recommended improvements for calculating ECPs. The ECP
enhancements were to: (1) revise certain constants used by the core design computer
models; and (2) use the unbiased boron concentration for the hot full power (HFP) and hot
_
O
.
23
,
zero power (HZP) conditions. The "best estimate" critical boron concentration includes a
boron bias which is used to correct for inaccuracies in the boron concentration predicted
by the core design computer codes. These recommendations were also provided to site
Reactor Engineering for review (Memorandum, " Estimated Critical Position Calculations for
MP 2 Cycle 12," dated April 28,1994).
The proposed calculation file and recommendations were not approved by the NFE
manager (Letter, " Proposed Calculation W2-517-405-NA and Associated Memo," dated
July 12,1994). The reasons stated were: (1) inadequate quality assurance review in
changing the fuel vendor's core design computer model; and (2) the desire not to bias ECP
predictions using past ECP errors. The NFE manager recommended an effort be initiated to
enhance the fuel vendors core design models. Reactor Engineering also concluded that the
recommended changes could not be used to improve the ECPs (Memorandum, " Estimated
Critical Position Calculations for MP2 Cycle 12," dated August 16,1994).
The licensee fuels engineers efforts to improve the ECP accuracy continued throughout
1994-1995. On November 15,1995, the fuel vendor provided a revised Startup and
Operations Report (" Transmittal of Millstone Unit 2, Cycle 13, Startup and Operation
Report, EMF-94-201(P), Rev.1 and Updated XTGPWR Deck") which used the latest
neutronics design methodology. The primary improvements were: (1) finer depletion steps
for the gadolinium cross sections; and (2) more accurately reflecting the full power fuel
temperatures. The methodology also reflected, to a lesser extent, the core design code
improvement recommended in calculation W2-517-405-NA. A significant improvement in
the accuracy of the ECPs was demonstrated using the revised analysis,
c.
Conclusions
The cycle 12 and 13 ECPs reviewed complied with technical specification and
administrative requirements. The licensee's corrective action to improve the ECPs by
improving the core design codes was a technically sound approach to resolve this issue.
The basis for rejection of the recommended ECP methodology changes using the unbiased
boron concentration was appropriately documented and justified. The inspector concluded
that the actions taken to improve the ECP calculation accuracy were appropriate.
E2.3
Boron Biases
a.
Inspection Scope (37550)
The licensee identified a concern that the boron biases may have an adverse effect on the -
core safety analysis. The inspector reviewed the actions taken to evaluate and resolve this
concern.
The "best estimate" boron concentration, used in certain core safety analyses and startup
physics testing, includad a boron bias. The boron bias was the difference between the
measured hot full power (HFP) boron concentration and the predicted boron concentration
as calculated during past cycles. The predicted boron concentration was calculated using
computer core design models. The "best estimate" boron concentration was equal to the
sum of the boron bias and the predicted boron concentration. The boron bias was added
to compensate for consistent inexactness in the computer core design models.
_. _.
.
.
.-
-,
.
--_
.
-
.
-
.-_
.
.
4
-
24
b.
Observations and Fin.dinas
The magnitude of the biases used in the prediction of HZP and HFP critical boron
concentrations for cycle 13 ranged from 55 to 80 parts per million (ppm) boron. A Nuclear
,
Fuel Section (NFS) Engineer outlined several safety concerns regarding the boron biases
(Memorandum, " Concern of Large Biases in Predicted MP 2 Critical Boron Concentrations,"
dated September 21,1995). The primary concerns were that: (1) the boron biases may
have an adverse effect on the plant safety analysis; and (2) the acceptance criteria used in
the startup physics testing, which uses the "best estimate" values, appears to be
inconsistent with the users guide provided as Appendix A to American National Standards
.
Institute /American Nuclear Society (ANSI /ANS) standard 19.6.1-1985, " Reload Startup
,
i
Physics Tests for Pressurized Water Reactors."
The NFS Supervisor provided a preliminary response to these concerns (Memorandum,
" Preliminary Evaluation - Large Boron Biases at MP2," dated October 5,1995). The
preliminary response stated that the main impact on the safety analysis would be the boron
concentrations used to analyze the boron dilution events and for shutdown margin
2
calculations. The fuel vendor confirmed that the same boron biases are applied to the
'
safety analysis calculations. The other potential effect boron biases could have on the
safety analysis was a perturbation of the radial power distribution. The evaluation
concluded that the affect on the radial power distribution was not large enough to be a
safety concern.
q
The preliminary response also addressed the concern with the apparent deviation from the
ANSl/ANS standard. The resporse stated that Millstone Unit 2 used the standard as a
general guideline, but were not committed to conduct core physics testing in accordance
j
with the standard. They also noted that the recommendation to use the unbiased boron
concentrations for physics testing was provided in the optional part of the standard. The
evaluation stated that in an upcoming revision to the standard, a current proposal is to
reverse this position and use the "best estimate" boron concentrations for comparisons
during physics testing. The response stated that the fuel vendor recommended using the
"best estimate" values for the core physics testing (Letter, Millstone Reactivity Biases,
,
i
dated October 2,1995).
An evaluation of the effect of the boron bias on the safety analysis was conducted by the
,
engineer who originally identified this concern (Memorandum, " Review of the MP 2 Cycle
13 Safety Related Analyses," dated Decemt er 12,1995). The evaluation assessed the
- '
effect of the boron bias on power distribution affected parameters, shutdown boron
concentration, and the boron dilution transient analysis. The conclusion of this evaluation
'
was that using the best estimate boron concentration for certain safety analyses and for
conservatism in the shutdown boron concentration adequately compensate for the
l
inaccuracy in the predicted boron concentrations. The conclusion was that the inclusion of
the boron bias did not result in significant changes in safety-related parameters. The
overall conclusion was that the results of the cycle 13 safety analyses and the shutdown
!
boron concentrations remain valid.
,
.
.
25
c.
. Conclusions
The inspector concluded that the licensee had conducted a thorough evaluation of this
concern. The evaluations demonstrated that the safety analyses for cycle 13 were not
adversely affected by the inclusion of boron biases. The basis provided for using of the
"best estimate" boron concentrations as the predicted values for core physics testing was
i
acceptable. The detail and timeliness of the evaluations of this concern were
commensurate with the potential safety significance of this issue.
U2.E8
Miscellaneous Engineering issues
E8.1
(Closed) LER 50-336/95-19: Shutdown Coolina System Inocerable due to
Damaaed Snubber Sucoort
a.
Inspection Scope
The inspector evaluated the licensee's disposition of failed snubbers on the suction header
of the facility 1 emergency core cooling system (ECCS) pumps,
b.
Findinas and Observations
On May 14,1995, with the unit shut down, a hydraulic snubber support assembly on the
facility 1 ECCS suction header was determined to be inoperable due to a significantly bent
extension rod. In addition, the hydraulic snubber had rotated on its axis causing its
hydraulic fluid supply reservoir to be located below the valve assembly. On May 11,
1995, a trouble report had been written to address that hydraulic fluid had been observed
leaking from the vent port. The shutdown cooling (SDC) system was declared inoperable
after 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> had passed without evaluation of the deformed snubber assembly. The
SDC system was appropriately declared inoperable, however, the system remained in
operation to remove decay heat from the reactor.
The licensee event report (LER) stated that the root cause of the event was being
investigated to determine the origin of the load that bent the extension rod. The licensee
was investigating both water hammer and external loads as potential causes. The initial
corrective action was to repair the snubber assembly to restore SDC system operability.
The LER stated that additional corrective action will be determined based on the results of
the root cause investigation that was underway and that this would be reported in a
supplement to the LER.
The inspector had three concerns with the LER: (1) The licensee has yet to meet their
commitment of submitting a supplement to the LER to discuss the results of the root cause
investigation; (2) The root cause investigation has not been completed even though the
event occurred more than 14 months ago; and (3) the LER did not specify a date that their
planned corrective action would be completed.
The licensee stated that the root cause investigation is ongoing because they have been
unsuccessful in definitively determining the cause of the damaged snubber. Their
investigation revealed that in addition to the hydraulic snubber, a mechanical snubber on
1
.
i
.
26
the same line was also damaged. Evaluations of various scenarios such as starting pumps
and operating valves showed that the resulting water hammer forces would be insufficient
to cause the observed damage to the two snubbers. The inspect- informed plant
engineering that operators had related that when the motor-operated valve at the RWST is
opened to fill the ECCS suction header, it " rattles the roof," indicating the magnitude of the
water hammer. The licensee stated that system fill was one of 7 scenarios considered but
was not pursued because of engineering judgement that the system fill would not provide
sufficient force. Plant engineering stated that they met with operations personnel and
there was no mention of the significant water hammer that has occurred during previous
ECCS suction header filling. Also, a more recent adverse condition report addresses a
piping support base plate near the RWST that had two bolts pulled out from the wall. This
provides additional evidence of significant transients in this system.
c.
Conclusions
At the end of the inspection period, the licensee had not yet completed calculations to
confirm whether the system fill could be the cause of damaged snubbers. They also plan
to perform detailed walkdowns of piping supports of both ECCS suction headers to
evaluate the current condition of the supports in the ECCS suction piping. The timely
completion of the licensee's evaluation of the current system status, the determination of
the root cause, and implementation of corrective actions are important due to the potential
for inoperable supports to render all ECCS pumps in both trains inoperable. Resolution of
the water hammer issues, as well as the concerns associated with the LER commitments
are considered unresolved. (URI 336/96-06-08)
E8.2
(Closed) LER 50-336/95-43: Reactor Core Thermal Power Level Exceeds
License Limit
On November 15,1995, the reactor core thermal power level exceeded the maximum
power level permitted by the operating license (2700 megawatts thermal). The core heat
balance calculation had been performed using an incorrect value for steam generator
blowdown flow rate. This resulted in the calculated core thermal power being less than
the actual core thermal power. The license limit was exceeded for approximately 11
hours. A best estimate of the maximum steady-state power level achieved during this
period was 2709 megawatts thermal (approximately 100.33 percent power). The
inspector reviewed computer records to verify that the licensee's immediate corrective
action to input an acceptable blowdown flow rate value into the calculation had been
completed. This issue is considered unresolved pending further review of the final
resolution of blowdown flow input, and evaluation of the control and validation of plant
computer calculations. (URI 336/96-06-09)
.
.
27
E8.3
(Closed) LER 50-336/96-06: Service Water Pumo Desian Vulnerable to Flood
Water
Technical Specifications require that one service water pump be protected from flood
waters when severe storm conditions threaten. This assures that at least one pump will be
operable for use after the flood conditions subside. The Unit 2 service water system flood
protection design provides protection for only the "B" or "C" service water pump. The
licensee determined that there had been outage periods when neither the "B" nor "C"
pump remained operable such that they could have been protected for use after a flood.
During these periods, however, no severe weather conditions were experienced. In
response to this design deficiency, the licensee implemented administrative controls to
assure that the "A" service water pump is not used as a single operable pump. The
inspector verified that procedures OP 2264, " Conduct of Outages" and OP 2326A,
" Service Water System," were changed in March and July 1996, respectively, to formalize
these corrective actions. Operators received training on the new controls through required
reading.
The inspector determined that the service water flood protection design requirements had
not been correctly translated into specifications and procedures. This is a violation of 10
CFR 50 Appendix B, Criterion Ill, " Design Control." This licensee-identified and corrected
violation is being treated as a Non-Cited Violation, consistent with Section Vil.B.1 of the
E8.4
(Closed) LER 50-336/96-08: Containment Sumo Screen Mesh has Holes
Laraer than Desianed.
a.
Inspection Scope (92700)
On February 20,1996, Unit 2 was shutdown due to concerns that small post-accident
debris, which passes through the containment sump screens, could clog the small
openings in the high pressure safety injection (HPSI) system throttle valves. Subsequent
licensee inspection of the sump screens revealed many holes in the screen mesh that
exceeded the design mesh size. These concerns were reported to the NRC in LER 50-
336/96-08. The inspector reviewed the identified problems and verified the safety
consequences and corrective actions.
b.
Observations and Findinas
The containment sump screens are designed to prevent debris from clogging the
containment spray and emergency core cooling systems (ECCSs) during the sump
recirculation phase of an accident. Failure to meet this design intent raised a serious
potential for common cause failures of these safety systems. Licensee review of an
industry operating experience report led them to the fact that the screen mesh size (0.187
in') could pass materiallarger than the throttled opening of the HPSI throttle valves. The
industry experience report mitigated the significance of this concern because the low
pressure safety injection (LPSI) system provides the primary cooling source in the
recirculation mode. Since HPSIis the only system used for the recirculation phase of
emergency cooling at Unit 2, the plant was shut down while analysis and corrective
.
.
I
28
actions were pursued. Licensee analysis subsequently showed that containment sump
flow velocities were sufficiently low such that only low density materials could be
transported from the screens, up to the pump suction pipe openings. These openings are
11 inches above the containment floor. The licensee concluded that any low density
material that reached the suction pipe would be pulverized in the close tolerance safety
injection pumps, and/or would be forced through any smaller orifice by high system
differential pressure at these points. The inspectcr was concerned that the suction strainer
had an open mesh top. High density material sinking into the water in containment could
possibly flow into the suction pipes through the top grating. As described below, the
licensee redesigned and fabricated a new sump strainer assembly. The new design
incorporated a solid top, thus resolving this NRC concern.
I
Unit 3 also reassessed the potential for containment sump debris clogging ECCSs during
accidents. The licensee's operability determination similarly concluded that only low
density material would transport, and it would not prevent fulfillment of the safety
function. Initial NRC review of this concern did not identify any immediate safety
concerns. However, further licensee review raised questions regarding the need for
operator action to compensate for clogging or accelerated erosion due to high flow at these
choke points. The issue of low density material smaller than the screen mesh design
effecting ECCSs remains unresclved pending NRC review of any required operator actions,
and confirmation that the harder debris cannot be transported to the pump suction (URI
336/96-06-10).
During the February 1996 Unit 2 shutdown, the licensee inspected and compared the
l
containtnent sump strainer against the current design specifications. Several discrepancies
I
were identified where debris much larger than the screen mesh size could pass through the
I
strainer. Specifically, the two end panels and the center partition of the strainer were
l
constructed of wire mesh with greater than the designed (0.187 in') openings. In addition,
there were ten locations where openings as large as 0.25 inch by 2 feet were identified.
The cause of the strainer discrepancies was construction / installation error and poor
oversight. The strainer was last worked on in January 1988 when the center partition was
noted to be missing. However, repair efforts did not assure that the correct screen mesh
size was installed at that time, nor did licensee response to this discrepancy identify the
other construction / design discrepancies which apparently existed at that time.
Technical Specification 3.5.2, "ECCS Subsysterns," and 3.6.2.1, " Containment Spray
System," require two operable trains of ECCS and containment spray during plant
operation at power. Because the sump strainer would pass debris larger than the system
design, the potential to compromise the safety function of both trains of ECCS and
containment spray existed. The low density material size would be reduced passing
through ECCS pumps and differential pressure at choke points would tend to pass this
material through. However, this position did not address the potential for higher density
material to reach ECCS components from the top of the st ainer and could not confirm the
operability of the ECCSs during prior operations with a strainer that would not perform its
l
design function. While the probability of a loss of coolant accident with loss of ECCS
function is !ow, the consequences of that scenario are unacceptable, and must be
prevented.
I
l
)
!
-
,
.
!
29
The licensee redesigned the strainer to resolve the discrepancies. These repairs were
implemented using the recently improved design control process that provides stronger
j
controls than those in place when the strainer was last modified. Also, the licensee is
engaged in a comprehensive verification of the plant design basis, which should identify
any other significant system design flaws.
,
i
In February 1996, the licensee also reported another potential ECCS analysis discrepancy.
Review of the containment sump recirculation design basis had revealed that reactor water
storage tank level could decrease faster than the final safety analysis report (FSAR)
'
described scenario. Therefore, it was questioned whether HPSI alone could cool the core
at this time, since LPSI shuts down during the sump recirculation mode. The licensee
i
subsequently determined that adequate HPSI cooling exists at the earlier switchover time
i
and retracted the report. The inspector reviewed the licensee's justification and had no
further questions regarding the adequacy of sump recirculation timing. The licensee will
verify the adequacy of the FSAR description of this function during their ongoing design
,
basis review.
)
,
,
,
c.
Conclusions
10 CFR 59, Appendix B, Criterion XVI requires conditions adverse to quality such as
,
i
deficiencies to be promptly identified and corrected. The containment sump strainer
deficiencies represent a potential common mode failure that could have rendered both
ECCS and containment spray inoperable. The licensee's identification of this problem
based on the review of recent industry operating experience demonstrated a current
i
conservative approach to safe plant operation. However, the failure to address these
issues when other discrepancier were identified in 1988 represents a failure of the
3
licensee's corrective action prog'am. These are apparent violations of TSs 3.5.2 and
3.6.2.1, and Criterion XVI. (eel 336/96-06-11)
!
l
E8.5
(Closed) LER 50-336/96-13: Potential Common-Mode Failure in Wide Ranae
Nuclear Instruments
_
,
This event involved the discovery of a potential susceptibility to common-mode failure
within the wide range nuclear instrument (WR-NI) channels. The problem is caused by the
presence of one nonsafety-related annunciator circuit that interfaces with all four channels
i
,
of the WR-Ni instrumentation. The annunciator circuit was designed with coil-to-contact,
'
to contactor isolation. However, the WR-Nis experienced cross-channel interference
through the common circuit on March 8,1996, caused by a single power supply
,
"
malfunction. The inspector noted that the LER did not discuss permanent modifications to
resolve the design concerns. Those actions are detailed in NRC Inspection Report 50-
336/96-201.
The LER also noted that the original WR-NI reactor protection trip on high rate-of-change in
,
.
power had been removed in 1978. The licensee subsequently determined that the removal
l
of this reactor trip may not have been consistent with the current methodology for analysis
1
of a rod withdrawal accident. This problem was promptly reported to the NRC on July 17,
]
1996 and supplemented on August 12,1996. NRC will review the cause and corrective
actions for this issue upon receipt of the followup LER to these telephonic reports.
'
__ _ _ ___
.
P
.
30
E8.6
(Closed) LER 50-336/96-19: Electrical Eauipment Qualification of Solenoid
Operated Valves inside Containment
a.
Inspection Scone
The inspector evaluated the licensee's response to LER 50-336/96-19.
b.
Observations and Findinas
On March 26,1996, the licensee discovered that the connectors for seven solenoid
operated valves located inside containment did not have the requirei electrical equipment
qualification (EEO) for operation in a harsh environment. They had previously determined
that the connectors for these containment isolation valves were not required to be
qualified. This was based on an incorrect assumption that the valves' only safety function
was to close and a harsh environment would cause a short resulting in the valves fai!ing
closed on a loss of power. During recent efforts to reorganize EEQ databases, the licensee
recognized the mis-identified safety functions of these EEQ components. In fact, it is
necessary +o reenergize these solenoids to open the valves later in the post-accident
scenario. The affected components included the containment air radiation monitors,
hydrogen monitors, post-accident sampling system, charging supply, pressurizer auxiliary
spray line, and hydrogen purge valves.
The LER stated that the root cause of this event was that programmatically, the licensee
had not completed an adequate review to define all the safety functions that individual
components and circuits must perform and the duration over which they must perform that
function. This weakness had been previously identified and in 1993 the licensee created
an EEQ Program Manual that formally delineated responsibilities of key groups that provide
input into the EEQ program. Safety Integration and Analysis (Sl&A) was defined as the
responsible group for providing the safety functions and operating durations of EEQ
equipment. However, Sl&A did not begin their reassessment until 1995, and are not yet
completed, in LER 96-19, Unit 2 committed to complete the process of redefining the
safety functions of all EEQ components and to disposition identified deficiencies prior to
entering Mode 2.
Prior to Mode 4, the licensee committed to correct the seven solenoid valves connectors
that were identified and to update the associated EEQ documentation. NRC interviews
with the Unit 2 coordinator for the EEQ Program revealed that although there may be EEQ
documentation for individual components, there is no EEQ documentation for the circuit as
a whole, which includes connectors. The licensee is in the process of preparing the EEQ
circuit documentation.
The non-EEQ connectors for four of the seven solenoid valves had previously been
discussed at a 1988 enforcement conference (NRC Inspection Report 50-336/88-20).
These included the containment isolation valves for the containment air radiation monitors,
the hydrogen monitors, the post-accident sampling system, and the hydrogen purge valves.
The proposed violation discussed 10 solenoid operated valves that did not did not have
EEQ connectors. At the enforcement conference, the licersee provided reasons that they
believed that enforcement action was not warranted. The licensee stated that only one of
-
.
.= -
-
-.
.
.
-
.
- -.
1
l
.
31
the 10 valves, an atmospheric dump valve, had terminations that required environmental
qualification because this was the only valve that required energization to perform its
j
safety function (i.e., valve opening). The other nine valves, all feedwater or containment
isolation valves, were said to go to their safe (i.e., closed) positions when their coils are
deenergized. This information was incorrect for the four containment isolation valves
because these valves also had a safety-related function to open. As a result, the NRC did
'
not consider the safety consequences associated with the four unqualified solenoid valves
when making enforcement decisions. In addition, the inaccurate information prevented
implementation of corrective actions to replace the unqualified connector 6 at that time.
The inspector reviewed the licensee's EEQ program more broadly and found that three
1
elements involved in an effective program have not been adequately completed to date,
including: (1) As stated above, the safety function (s) of individual components and the
duration they must perform that function has not been comprehensively and accurately
defined for EEQ program components; (2) The licensee has determined that the 1986 EEQ
field walkdowns were inadequate because the walkdowns were not comprehensive; the
walkdowns were performed by maintenance technicians and contractor personnel that
were not adequately trained; and many concerns identified in the walkdowns were not
adequately dispositioned. As a result, during the current shutdown, the licensee plans to
complete a comprehensive walkdown of accessible EEQ equipment; and (3) The licensee
determined that EEQ preventive maintenance requirements had not been incorporated into
the Unit 2 maintenance tracking system.
Although not applicable to the solenoid operated valves discussed in this LER, the
inspector also noted that the licensee has not yet completed their High Energy Line Break
(HELB) Program for Unit 2. This project assesses the harsh environment parameters
(temperature, pressure, humidity, radiation levels, etc.) for spaces outside containment. In
the licensee's "Short Term Review of Final Safety Analysis Report Amendment 23 (HELB
effects)," dated August 1,1990, the licensee stated that, " engineering has been performed
and plant modifications installed without regard to impact upon the High Energy Line Break
program." The licensee's initial corrective actions included performing a "short term" or
" cursory" reevaluation of the 1973 HELB report to look for " obvious programmatic
problems." This short-term review identified the need to redefine realistic subcompartment
environmental conditions because "it became increasingly apparent that environmental
conditions identified in a number of plant areas were inaccurate." In 1990, the licensee
discovered that the environmental conditions associated with an auxiliary steam line break
were much more severe than the existing design basis break postulated in the main steam
system. Walkdowns were performed in areas of the plant that were considered mild
environments, but where auxiliary steam piping penetrated, to determine what safety-
related equipment could be affected by a postulated auxiliary steam line break. The areas
affected were the control room, the control room ventilation room, and the auxiliary
building 14' 6" elevation. Discussions with the licensee indicated that although this
concern was identified in 1990, the monetary and personnel resources necessary to
redefine the environmental conditions have routinely been diverted such that this project
has not yet been completed. Similarly, the licensee also has not completed the project to
evaluate the dynamic (pipe whip, steam impingement) consequences of a pipe break.
-
_
_ .
.
.
__
.
.
--.
__ ..
.
-
,
I
32
c.
Conclusion
i
10 CFR 50.49 requires that: (1) each item of electric equipment important to safety shall
'
be qualified by testing and/or analysis of identical or similar equipment, and the
qualification based on similarity shallinclude a supporting analysis to show that the
equipment to be qualified is acceptable; and (2) a record of the qualification shall be
maintained in an auditable form to permit verification that each item of electrical equipment
important to safety is qualified and that the equipment meets the specified performance
i
requirements under postulated environmental conditions. The faibre to adequately
,
l
establish the qualification of the connectors for the seven solenoid valves discussed in LER
50-336/96-19 is an apparent violation (eel 336/96-06-12). This is of particular concern
,
because four of the seven valves were the subject of previous escalated enforcement
activities in 1988. Due to inadequate licensee reviews, inaccurate information was
provided to the NRC regarding the safety function of the valves and therefore, this safety
concern was not properly dispositioned.
Programmatically, the failure to accurately define the harsh environment parameters for
equipment outside containment and to correctly define the safety functions of all EEQ
'
equipment is a significant concern because this information provides the foundation for
,
qualification of each component. When combined with an incomplete understanding of -
l
what is currently installed and its condition, this raises the uncertainty regarding the ability
l
of safety related components to perform their design function (s) in the environment the
j
component or circuit may experience. Although the EEQ Program and HELB Program are
within the scope of the 50.54(f) design reviews that are currently underway, significant
licensee management focus in this area is needed to support restart of Unit 2. In addition,
since the same organization implemented the original EEQ program requirements for all the
licensee's nuclear units, it is likely that similar problems may exist at the other units.
l
i
. .
.=
. _
._.
.
- .
..
.
.
33
Report Details
l
Summary of Unit 3 Status
Unit 3 remained in cold shutdown throughout the inspection period. The licensee's review
to verify compliance with their established design and licensing basis is ongoing.
U3.1 Operations
U3.01
Conduct of Operations
01.1
General Comments (71707)
i
The inspectors conducted frequent reviews of ongoing plant operations; including control
room activities, unit daily status meetings, management review team (MRT) evaluations of
adverse condition reports (ACRs), and assessments of operability determinations (ODs) and
reportability evaluations for degraded plant conditions. Discussions with licensed plant
operators and management personnel revealed good cognizance of the current plant
conditions and expected plant problem areas. For example, during the current outage, it
was anticipated that both shutdown margin monitors (SMMs) would become inoperable
due to the decay of the neutron sources that provide a minimum count rate to keep the
SMMs on scale. NRC Inspection Report 50-423/96-05 provides a detailed discussion of
this issue, as documented in ACR 12495. On June 21,1996, the licensee issued a
contingency action plan, approved by the plant operations review committee (PORC), for
inoperable SMMs. Hence, when the second of the two SMMs went off scale and was
declared inoperable on July 25,1996, the licensee had in place the appropriate action plan
to meet the requirements of the technical specification (TS) 3.3.1, action 5(b).
The inspector reviewed the implementation of other compensatory measures; e.g., the
issuance of Bypass / Jumper 3-96-076 to address a deficiency documented in ACR M3-96-
0568 involving an electrical manhole cover, tornado design restraints. The ODs and
reportability evaluations for additional ACRs were also reviewed. The inspector attended a
PORC meeting on July 26,1996, and observed a good questioning attitude by PORC
members in evaluating the adequacy of a new project instruction intended for issuance as
part of the Unit 3 Configuration Management Plan (CMP). Subsequently, the resident
inspectors were apprised of licensee Nuclear Safety and Oversight assessment activities
relating to the ongoing engineering reviews involved with the CMP. Overall, the licensee's
approach to problem identification and resolution (e.g., ACRs) and process controls (e.g.,
the CMP) appeared to be receiving an appropriate level of management attention and
oversight.
Additionally, using inspection Procedure 71707, the inspectors observed various routine
plant evolutions and norrral shutdown operational activities to verify the acceptability of
the overall conduct of operations. With respect to operational controls, particularly in
consideration of shutdown risk criteria, Unit 3 was found to be operated safely in cold
l
shutdown (mode 5) conditions during this inspection period.
. .. ..
- _ - .
_
-
- , .
. - __.-. - -=
. - - - - - _
_-
.
.
~.
j
i
34
l
i
j
U3.07
Guality Assurance in Operations
07.1
Audit of the Quality Assurance Proaram (40500)
(Ocen) IFl 50-423/96-06-xx
.
j
During the inspection period, the inspector attended the June 28,1996, Joint Utility
Management Assessment (JUMA) team exit. TN JUMA audit was to evaluate the
]
effectiveness of the licensee's Quality Assur-
TA) Program. The scope of the audit
was to evaluate actions taken to address pre,
., assessments of the QA organization,
-
evaluate philosophy, guidance, and staff understanding for the designation of critical
attributes for work activities, review the implementation and effectiveness of the audit and
surveillance programs, and to assess the value added as a result of QA activities.
.
l
The JUMA team concluded that the audit, surveillance, and inspection programs at
Millstone were not effective in the implementation of their Mission Statement and the
resolution of identified problems. The team attributed these problems to:
e
Lack of support for the QA organization by executive and line management.
i
Lack of an effective corrective action program.
1
Some adverse condition reports were generated as a result of the audit findings. One
addressed that there were no requirements to respond to audit findings within 30 days as
required by ANSI /ASME N45.2.12, " Requirements for Auditing of Quality Assurance
Programs for Nuclear Power Plants," and another identified ti.at the Unit 1 Measuring and
Test Equipment audit had not been performed within the specified 24 months. The JUMA
audit also identified that the designation of critical attributes for work activities was wed.
The inspector concluded that this assessment activity was effective in identifying
significant problems in the QA program. At the conclusion of this inspection, the licensee
was developing an action plan to address the findings in the JUMA audit. The licensee
indicated that the corrective actions would be folded into their Nuclear Excellence Plan.
Actions taken to address these concerns are considered an item for further inspection
followup and will be addressed in the Restart Assessment Plan.
U3.08
Miscellaneous Operations issues (92700)
08.1
(Closed) LER 50-423/96-11: Surveillance testing revealed that both trains of
the control room envelope pressurization system (CREPS) were inoperable in
violation of Technical Specification (TS) 3.7.8. The 36 foot elevation of the
control room was unable to achieve the required positive one-eighth inch
differential pressure due to an imbalance in the air-conditioning system that
serves the control room. The imbalance was determined to have occurred 17
days earlier after modifications were made to the control room. The inspector
reviewed the Final Safety Analysis Report and found no detailed description of
__ _
-
-
-
- -
--
_
- -
..
.
.
- -
--
._
-
.
.
35
the functional relationship between the control room air conditioning and
CREPS. Therefore, the adverse impact of the imbalance was not considered
until the . surveillance testing identified the problem. This licensee identified
and cor scted violation is being treated as a Non-Cited Violation, consistent
with Section Vll.B.1 of the NRC Enforcement Poliev. This LER is closed.
08.2
(Closed) LER 50-423/96 16: This LER documented a condition outside the
design basis of the plant and an inadvertent Engineered Safety Feature
actuation. The safety related 4160 volt switchgear cabinet seismic
qualification had not been maintained as a result of the bolts on the rear door
and the seismic latches on the front door not being used. While engaging the
latches to restore the seismic qualification, a relay mounted on the door
i
actuated, resulting in a control building isolation. The inspector examined the
switchgear door and latch configuration and discussed the identified concerns
j
with engineering perscnnel. This LER documented an issue with minor
j
consequence and only hypothetical significance. Corrective action was
effected and this LER is closed.
l
l
l
08.3
(Closed) URI 50-423/96-04-12. Unanalyzed Containment Pioina Desian:
(Closed) LER 50-423/96-07. Containment Recirculation Sorav and Quench
Sorav System Outside Desian Basis
a.
Insoection Scope (92901)
On April 3,1996, the licensee determined that the plant had operated in a condition that
was outside the design basis due to a deficiency in the design of the recirculation spray
i
system (RSS) piping supports, for which the loading analysis had not appropriately
l
considered accident temperatures. Subsequently, the licensee determirad that based upon
design basis act:ident temperatures inside containment, the unacceptable pipe support
,
I
stress conditions also applied to the quench spray system (OSS). LER 423/96-07 was
submitted on May 2,1996, to document these deficient conditions. Unresolved item (URI
l
423/96-04-12) was documented during a follow-up NRC inspection to track the licensee's
continued engineering analyses and corrective actions. In NRC Inspection Report 50-
i
i
I
423/96-05, further review of the design change and modification work packages affecting
the RSS pipe supports was documented. During this inspection, the inspector assessed
the status of the continuing design reviews, the field work and component modifications,
and the overall implementation of corrective measures.
b.
Observations and Findinas
The inspector reviewed completed work packages and noted that the RSS and OSS
modifications have been completed. However, the licensee determined that other systems
[i.e., safety injection (SI), both inside and outside containment, and the reactor plant
component cooling (CCP)] require similar analysis. The pipe support and structural steel
reviews for these other systems resulted in the need for additional pipe support and
structurai modifications. Installation of these design changes has been ongoing during this
inspection period.
-
_-
- .. -
. . . - - -
_ .
_ - - - - = _ .
--
-
-
_ . .
- -
.
.
36
The inspector also reviewed several engineering records to assess licensee actions relative
to the origin of the RSS design concerns. In 1993, the licensee identified RSS Design
l
Basis Documentation Package (DBDP) discrepancies that led to the documentation of
l
adverse condition report (ACR) 159 in 1995. The licensee's followup of this adverse
condition in March,1996, led to the identification of the current containment temperature
concerns wherein post LOCA containment temperatures cause excessive piping system
stresses. While the initial review of these concerns was documented with the issuance of
ACR 10773, several other ACRs have since been issued. These document related
problems with elevated containment temperature effects and single failures that could be
postulated to raise the temperature of system f!uids beyond that which had been
previously analyzed. The inspector confirmed that such issues, and related problem
examples, have been considered in the report of the project instruction (PI) 2 team
findings, relative to the Unit 3 specific assessment conducted as part of the licensee's
Configuration Management Plan.
Notwithstanding the generic reviews and corrective measures implemented by the licensee
since the issuance of ACR 10773, LER 423/96-07 documents the fact that the RSS and
OSS systems were found to be in noncompliance with design-basis requirements. The
licensee's basis for initial reasonable assurance of continued operability for both systems
with the plant in mode 4 and heading to cold shutdown (mode 5) conditions suggested
that neither the RSS, nor the OSS system could be considered operable under full power
operating conditions. This design deficiency, adversely affecting the operational status of
systems required to mitigate the consequences of an accident and governed by the unit
technical specifications, has existed since the issuance of the initial operating license for
unit 3 in 1985. While the licensee has committed to submit a supplement to LER 423/96-
07 by September 13,1996, addressing the generic implications to other p! ant systems, the
inspector noted that the need to modify certain Si and CCP pipe supports has already been
established and is in progress.
c.
Conclusions
With respect to the design-basis capability of the RSS J.nd OSS system functions, Unit 3
}
was operated in violation of the Unit 3 technical specifications over a period of several
years. Furthermore, since at least 1993, a DBDP deficiency documented concerns in this
area, but was not adequately addressed by the licensee's corrective action program until
ACR 10773 was initiated in 1996. While the licensee plans to submit an LER supplement
to update the generic implications and status of activities related to this problem, the
existing facts support the position that past operation of Unit 3 with this design deficiency
is an apparent violation (eel 423/96-06-13) of regulatory requirements.
U3.Il Maintenance
U3.M1
Conduct of Maintenance
!
M1.1
General Comments (62707)
On July 31,1996, the inspector attended the prejob briefing for maintenance activity M3-
95-609, loop calibration of containment recirculation pump 3RSS*P1 A discharge flow
.
__
. _ _
. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ - _ _ _ _ _ .
.
.
37
transmitter. The briefing included discussion of the adequacy of the tagout, potential
personnel safety hazards, and the required retest. The supervisor alerted the technicians
that the job would require a safety related component being declared inoperable, potentially
placing the plant in a technical specification action statement, and that blocking open the
access door to the room would affect a plant fire barrier. In addition, the supervisor
cautioned the workers to work within the job scope and that the identification of any
additional work would require that the work order be revised. The inspector concluded
'
that the brief by the instrument and control supervisor was thorough.
U3.M3
Maintenance Procedures and Documentation
M3.1
Trisodium Phosohate (TSP) Surveillance (61726. 92901)
a.
Inspection Scoce
Amendment No.115 to the Unit 3 operating license (OL) NPF-49 was issued on May 26,
1996. This revision, as implemented with plant design change record (PDCR) MP3-94-
135, amended the plant technical specifications, reflecting the replacement of sodium
hydroxide from the refueling water chemical addition tank (CAT) with TSP as the pH
control agent for the containment spray system. The inspector reviewed the licensee's
procedures and records relating to the TSP surveillance requirements, examined specific
field configurations of the completed PDCR, and assessed the impact of the resulting
modifications upon the Unit 3 Final Safety Analysis Report (FSAR).
b.
Observations and Findinas
The inspector reviewed Surveillance Procedure (SP) 3606.10 for the TSP storage basket
volume check, conducted at least once each refueling interval. Since a dodecahydrate
form of TSP was utilized in the design, a surveillance based upon a volume check
conservatively ensures a sufficient amount of TSP remains available for pH control of the
recirculation coolant during an accident. Even if any humidity-induced agglomeration
occurs, the density in the TSP baskets would increase; thus, providing for TSP additions to
the minimum refillline to effectively increase the mass of TSP available for accident
response. The inspector confirmed that the required surveillance was conducted during the
last refueling outage in May,1995, and repeated in April,1996. Adverse condition report
(ACR) 12327 was initiated on April 29,1996, to document the finding of the TSP below
the required fillline for all twelve containment baskets. This was not unexpected due to
the aforementioned agglomeration phenomenon. Nevertheless, for surveillance purposes,
corrective measures to refill all TSP baskets prior to taking the plant into mode 4 were
specified.
The inspector also examined the completed hardware modifications and pipe capping
activities, accomplished in accordance with automated work order (AWO) M3-95-00941,
that isolated the CAT from the refueling water storage tank. The applicable design
drawings and work records document the in-place abandonment of the CAT and associated
piping and components. Since the completed PDCR did not include the FSAR changes
relevant to PDCR MP3-94-135, the inspector requested the FSAR change transmittals
associated with this design change for further review.
I
1
- - - - .
_
-
-
_ - .
- _ - - - _ .
_ - _
_ _ _ .
.
.
.
38
l
Coincident with the NRC inspection of the records and FSAR change requests referenced
l
with the PDCR closeout, the licensee identified some discrepancies involved with the
incorporation of the PDCR design details and its safety evaluation (SE) into the Unit 3
l
FSAR. The deficiencies were found as part of the licensee's vertical slice review team
(VSRT) effort, conducted in accordance with project instruction (PI) 15, and were
,
documented in Unresolved item Report (UIR) log number 107. Most notably, the SE
j
assumed a containment leak rate greater than that noted in the current technical
specifications and in the OL Amendment No.115 revisions. Also, the FSAR chapter 15
accident analysis was found by the licensee to have not been reviewed for revision and
any impact relating to the TSP basket modification. As of the conclusion of this
inspection, the discrepancies identified on UIR log number 107 had been documented in
ACR 13788 and were pending final disposition. The inspector confirmed that Unit 3
Operational Readiness Plan Punchlist listed ACR 13788 as an issue requiring resolution
prior to the unit startup.
c.
Conclusions
The licensee's implementation of PDCR MP3-95-135 for replacement of the CAT with TSP
baskets inside containment was appropriately handled as a field modification and properly
controlled from operational and surveillance standpoints. However, with regard to the
licensing basis of this design change, discrepancies were identified that require further
analysis, a revision of the PDCR's documented safety evaluation, and additional changes to
the FSAR than were documented as part of the PDCR implementation. These issues are
currently being tracked by the licensee as items to be addressed prior to plant heatup to
mode 4 conditions.
U3.Ill Enaineerina
U3.E1
Conduct of Engineering
,
J
E1.1
Potential Cloaaina of the ECCS Throttle Valves - Uodate
i
a.
Insoection Scope (37551)
As documented in NRC Inspection Report 50-423/96-01, the licensee conducted an
operability determination (OD) to evaluate the concern [ reference: Adverse Condition
Report (ACR) 8897] that eight throttle valves in the emergency core cooling system (ECCS)
have openings smaller in size than the maximum dimension of the ECCS recirculation, fine
screen sizes. The potential for clogging the valves, and thus restricting ECCS flow during
the recirculation phase of safety injection, was analyzed in consideration of the nature of
the postulated debris, the " piggy-back" flow path through multiple pumps, and the
increasing differential pressure effects. While the licensee concluded that the affected
systems were operable at that time, additional information from other plants and other
phenomena (e.g., throttle valve erosion) were still being assessed for adverse system
impact.
.
._
..
.
i
.
39
b.
Observations and Findinas
The Millstone Unit 3 Final Safety Analysis Report (FSAR) documents plant compliance with
the regulatory position of NRC Regulatory Guide (RG) 1.82, regarding the ECCS sump and
containment spray design. Such guidance indicates that the recirculation sump screen size
openings should be based upon minimum restrictions downstream of the pumps and upon
recirculation system requirements. Also, Westinghouse Nuclear Safety Advisory Letter,
NSAL-96-001, discusses the identification of a potential problem involving the erosion of
the ECCS throttle valves under long term (i.e., post-accident, licensing-basis duration) flow
conditions. Also, a Westinghouse letter (NEU-91-611) identifies questions about the
adequacy of the runout tr' 4n for certain types of ECCS pumps.
The licensee intends to aos
.s both the NSAL concerns drid the RG 1.82 guidance with
system modifications. By installing orifice plates in the affected lines such that the
'
'
maximum pressure drop would not occur at the throttle valves, crosion problems would be
j
lessened and larger valve openings could accommodate the maximum debris size, as well
ds pump runout considerations. At the conclusion of this inspection period, the licensee
was still working on engineering provisions for the orifice plate installation and re-balanced
'
throttle valve settings. The high pressure safety injection (SlH) and charging system (CHS)
pump flow characteristics required further evaluation based upon the net positive suction
]
head (NPSH) boost from the " piggy-back" discharge flow from the recirculation spray
'
system (RSS) pumps.
Subsequently, the licensee determined that a combination of design factors, ii.cicding the
initial throttle ulve settings, the " piggy-back" flow path, the RSS discharge pressure
boost, and the . .
and SlH pump runout margins (i.e., the NEU-91-611 issue), required
documentation n,
OR M3-96-0524 and reporting to the NRC in accordance with 10 CFR
50.72 and 50.73. The licensee made the initial telephonic notification to the NRC
headquarters duty officer on August 30,1996, and is expected to submit an LER on this
issue within the next thirty days.
The inspector reviewed the reportability evaluation for ACR M3-96-0524 and determined
that the documented design considerations have relevance to the planned corrective
measures and engineering modifications for the throttle vaive clogging and erosion
concerns. The inspector also reviewed NSAL-96-001 and re-assessed the adequacy of the
OD for ACR 8897 in light of the new engineering issues that have been identified. The
inspector verified that the Final Safety Analysis Report has documented the specific Unit 3
differences between the plant configuration and either the regulatory guidance (e.g., RG
1.82) or the generic Westinghouse ECCS operational provisions; e.g., the RSS pump direct
flow path to the reactor vessel is isolated in favor of the " piggy-back" mode via the CHS
and SlH flow paths.
c.
Conclusions
'
The licensee's OD for ACR 8897 remains valid relative to the potential throttle valve
clogging concerns; however, consideration of licensing commitments to RG 1.82 and the
potential for longer-term valve erosion concerns require a plant modification to install
<
orifice plates in the susceptible ECCS flow lines. A recent licensee analysis identified a
_
\\
-
l
.
,
40
related concern (ACR M3-96-0524) with the potential for runout of the CHS and SlH
pumps during the recirculation phase of ECCS operation. This could have resulted in pump
cavitation with the loss of CHS and SlH pump functions during the " piggy-back" mode of
operation. Since this problem is also related to the initial settings of the ECCS throttle
valves, the licensee's design modification will have to consider such pump runout issues in
addition to the clogging and valve erosion concerns. Pending the licensee presentation of
evidence that the planned design change addresses allidentified problems and that the
submittal of a licensee event report to the NRC which documents the past deficiencies, this
issue is considered an unresolved item. (URI 50-423/96-06 14)
U3.E2
Engineering Support of Facilities and Equipment
E2.1
Adverse Condition Reoort (ACR) Review
a.
Inspection Scope (37550 and 40500)
The inspector reviewed selected ACRs to assess the effectiveness of the ACR process.
The evaluation included an assessment of the root cause determination and whether
appropriate corrective actions were identified and implemented to prevent recurrence of the
adverse condition.
l
b.
Observations and Findinas
ACR 06092 Reactor Coolant System (RCS) Valve Bodv-to-Bonnet Leak
j
On November 9,1995, a leak was identified coming from the "D" RCS loop accumulator
injection check valve at the body-to-bonnet joint. The valve is a Westinghouse ten-inch,
swing check valve. In taking immediate corrective action, the licensee entered the
applicable technical specification action statement, shut down the plant (reference NRC
Inspection Report 50-423/95-42), and generated an ACR to document and disposition the
problem. The ACR was categorized as a level "B" due to its consequence (reactor
shutdown) and its potential for recurrence. Therefore, a root cause investigation was
performed in accordance with nuclear group procedure NGP 3.15, " Root Cause Evaluation
Program." The inspector verified that the individual performing the evaluation had attended
the required root cause training and that the evaluation was documented in accordance
with procedure NGP 3.15 guidance.
The root cause investigation determined that the leak was attributed to a lack of metal to
metal contact (gap) on the leaking joint, as a result of the valve being reassembled
incorrectly in August,1993. The licensee inspected each installed Westinghouse check
valve for gaps and joint leakage and reviewed the maintenance history records for previous
leaks. This review resulted in the identification of eight valves (six and ten-inch
Westinghouse swing check valves) that had gasketed joints with questionable reliability.
Since disassembly of the these valves would have required placing the plant in mid-loop
condition, a plant design change request was instead processed to seal weld the body-to-
bonnet joints to ensure reliability of the body to bonnet seal.
j
.
.
41
As additional corrective action, the licensee plans to revise, by October 15,1996,
procedure MP 3766AH to require that the metal to metal fit of the body to bonnet joint be
verified when reassembling the valve. The licensee also is evaluating other options for
improving the reliability of the body to bonnet seal to eliminate the need to seal weld the
joint after further disassembles.
The inspector verified, by review of work orders, that the modification had been performed
for the eight Westinghouse check valves. A review of procedure MP 3766AH revealed
that the procedure change had not yet been completed. Action item requests had been
generated by the licensee to implement the remaining planned actions. No additional leaks
from Westinghouse swing check valves have been reported since November 1995.
ACR 1535 Temoerature excursion durina soent resin dewaterina
This issue was categorized as a level "A" ACR. It documented an event that occurred at
Unit 1 on June 22,1995. During the initial dewatering following the transfer of spent
resin from the spent resin tank (SRT) to the cask, the waste water temperature in the cask
liner rapidly rose from 90 F to 310 F. The inspector reviewed the licensee's investigation
and corrective actions.
The licensee's root cause investigation team was unable to recreate this event or
determine its specific cause. The licensee postulated that an exothermic reaction had
taken place in the liner, which resulted in the increased temperature. The licensee team
made several recommendations to mitigate the effects if a similar event were to recur. The
recommendations included: establish and maintain a constant line of communication
between the liner operator and the radwaste operator until drying begins, have flush water
available to refill the liner if heat-up occurs, use demineralized water for resin transfers, and
maintain the SRT full of water while the tank is unattended. These recommendations were
shared with the other Millstone units for implementation.
The inspector reviewed the applicable Unit 3 procedures and verified that all but the
recommendation involving a constant communication link had been properly
proceduralized. The implementing procedure required that a direct means of
communication be established, but it did not mandate continuous communications. The
licensee indicated that although the guidance did not specifically require constant
communications, such a practice has been implemented. The licensee indicated that the
procedure would be revised to include this recommendation.
ACR 1148 FSAR Not Uodated to Reflect New Site Buildina
This issue was categorized as a level "B" ACR. It documented that the external flooding
analysis presented in the Final Safety Analysis Report (FSAR) and an engineering
calculation had not been updated to reflect changes to the site. Four buildings had been
added or made permanent to the site between 1986 and 1993, without updating all
portions of the FSAR or updating pertinent documentation.
As corrective action, the licensee updated the design basis analysis calculations and the
FSAR to reflect as-built conditions. To prevent recurrence, the ACR recommended the
.
e
42
addition of an engineering representative to the site utilization committee and to the Unit 3
plant operations review committee (PORC).
The inspector verified that a FSAR change was processed, that the applicable engineering
calculation was performed, and that an engineering representative was assigned as a
member to the noted licensee committees. However, the inspector identified that the
engineering representative was not present at the site utilization committee meeting
convened in June 1996.
The inspector verified that the individual performing the evaluation for this ACR had
attended the required root cause training. However, a review of the root cause evaluation
revealed that it was not documented in accordance with procedure NGP 3.15 guidance. In
addition, the ACR was not reviewed by PORC as specified in the ACR. The licensee
generated ACR M3-96-304 to document and resolve these concerns.
j
The inspector determined that the licensee actions to correct similar types of problems
were questionable. There are no requirements that all committee members be present at
all meetings, nor are there procedural comrols in place to alert licensee personnel of a need
to consider the potential for flooding when erecting structures on site. The licensee
acknowledged the inspector's concern and planned to perform an additional review of the
issue. ACR M3-96-304 was modified to address this concern. The Events Analysis
department had not raised these problems during their closecut review of this issue.
,
l
The independent safety engineering group performed an evaluation of level "A" and "B"
'
ACRs to determine the effectiveness of corrective actions. The evaluation focused on the
timeliness and quality of root cause evaluations, the corrective action plans, and the
implementation and tracking of the corrective action plans. This review concluded that the
majority of the ACRs under review failed to meet one or more of the licensee's established
criteria. Only the immediate corrective action demonstrated an acceptable trend.
Furthermore, the inspector noted that an ACR had been previously written against the
corrective action program as a result of a Quality Assessment Service audit of level "C"
and "D" ACRs (reference NRC Inspection Report 50-423/96-05).
,
b.
Conclusions
The inspector concluded that the quality of the root cause investigations and corrective
actions associated with these ACRs was mixed. The root cause investigation for the
leaking check valve and the temperature excursion during the dewatering of the cask liner
was thorough. However, the corrective actions to address the failure to update the FSAR
only corrected the specific concern and would not have necessarily prevented recurrence.
The NRC expressed concern that none of the discrepancies had been identified by the
Events Analysis department, which has the responsibility to review ACRs for closeout.
The issue of an inadequate corrective action program has been previously discussed in
NRC inspection report 50-423/96-04. The NRC has indicated that prior to the startup of
any of the Millstone units, the corrective action program must be demonstrated to be
effective.
_
__
_
_
. _ _
-_ ____
.__
.__
e
43
4
E2.2
Loss of Foreian Material Exclusion (FME) Control
'
a.
Inspection Scope (71750)
On August 8,1996, while lifting the fuel handling tool from the spent fuel pool (SFP), the
4
bottom of the tool became caught on the lead-in for the transfer canal gate and fellinto the
SFP. The reactor engineer in charge of the evolution immediately stopped the job, notified
the shift manager, and generated an adverse condition report to document the event. The
j
inspector monitored the performance of the licensee's recovery efforts.
b.
Observations and Findinas
NRC Bulletin 96-01, " Control Rod Insertion Problems," documented a concern regarding
'
control rod binding problems in high burnup fuel at Westinghouse designed plants. As a
j
result of these concerns Millstone Unit 3, under the guidance of the Westinghouse Electric
Corporation, was one of several plants chosen to perform fuel testing and inspection of
4
j
selected fuel bundles ir, the spent fuel pool. After the completion of the required testing
and inspections, the Westinghouse representative was removing the aluminum fuel
i
handling tool when it caught and broke at a welded joint and fell to the bottom of the SFP.
'
Prior to the recovery efforts for the fuel handling tool, the reactor engineer reviewed the
root cause investigation for the Unit 1 SFP event (refer to section U1.M1.1) for lessons
,
learned, developed an action plan, and brief allinvolved parties. A work order was written
for retrieval of the foreign material.
The inspector reviewed the work order and verified that procedure WC-1, " Work Control
2
Process," guidance on recovery from loss of FME control was followed. Prior to retrieving
,
the tool, a camera was lowered into the pool to inspect the piece to determine / verify the
extent of the damage and determine the location of the foreign material. Indications
revealed that the tool was in two pieces next to the transfer gato. The inspector noted
-
that workers discussed each evolution in detail prior to its execution to ensure all parties
had a clear understanding of the actions to be taken and any potential consequences. The
workers implemented proper FME controls regarding logging materialinto and out of the
FME controlled area and demonstrated good radiological work practices.
c.
Conclusions
The inspector verified procedural compliance with the WC-1 guioance, and concluded that
the evolution of removing the foreign material was well planned and coordinated.
.
.
l
l
44
U3.E7
Quality Assurance in Engineering Activities
E7.1
Review of Desian and Confiauration Discrepancies
a.
Insoection Scope 192903)
j
In a letter dated June 20,1996, the licensee documented the Millstone Unit 3 Discrepancy
,
Review Team (DRT) Report. The report contained design and configuration management
l
deficiencies that were identified during licensee reviews, third party reviews, NRC
l
inspections or that were self disclosing through the occurrence of an event. An updated
j
!
report was provided to the NRC in a letter dated July 2,1996. The inspectors reviewed
the licensee's prioritization of th; deficiencies to ensure that issues deferred for resolution
after plant startup would not adversely affect the safe operation of the plant.
b.
Observations and Findinas
i
The July 2,1996, submittal indicated that as of June 25,1996, there were 1187 design
or configuration management issues identified. Of these, the licensee determined that 597
required resolution prior to plant startup. The remainder were scheduled for resolution by
either the end of 1996 or the next refueling outage. At the time of the inspection,
approximately 440 issues that were scheduled for resolution after plant startup remained
open.
The inspectors reviewed the summary description of each deferred item and selected for
further evaluation approximately 214 issues that appeared to be the most safety
significant. For these items, the inspectors reviewed the source document to obtain
additional details of the issues. The source documents included adverse condition reports
(ACRs), unresolved item reports (UIRs), open item reports (OIRs) and nonconformance
reports (NCRs). As a result of the review of the source documents, additional questions on
approximately 120 of the issues were directed to the licensee. During the resolution of the
questions raised by the inspectors, the licensee decided to revise the priority of 17 of the
issues and include them as issues to be resolved prior to startup. The resolution of these
items will generally involve procedure improvements, minor drawing changes, or the
generation of design documentation for minor modifications. Several of the issues will also
require an additional engineering review to ensure that the affected system operability is
not jeopardized.
c.
Conclusions
The inspectors concluded that the licensee's characterization of the issues was generally
appropriate. Most of the items that were upgraded to startup issues during the inspection
were associated with conflicting or missing documentation that would not have been
expected to have any significant impact on plant operation. The inspectors' initial
assessment was that the issues that require additional engineering review were not likely
to result in any safety significant problems. At the completion of this inspection, the
licensee evaluation process was continuing. Final assessment of the required work
prioritization is dependent upon the completed engineering reviews.
l
._ __
--
.
_ _ . __
.
_ _ .
- . - . - .
. . - -
-
_
.
45
U3.E8
Miscellaneous Engineering issues
E8.1
(Closed) URI 50-423/96-05-13. Desian Modifications for Letdown Heat
Exchanaer Leak Reoair
a.
Insoection Scope (92903)
As documented in NRC Inspection Report 50-423/96-05, the licensee was evaluating
certain NRC questions raised with respect to the implementation of plant design change
record (PDCR) MP3-90-243 for leak repair activities on the letdown heat exchanger. Of-
most significance was the issue of the ASME Code acceptability of the existing lower
flange bolted condition. During this inspection, the inspector reviewed the licensee's
responses to the questions on the PDCR package review and additional documentation
regarding the code compliance and adequacy of the design control measures associated
with this modification.
b.
Observations and Findinas
The inspector identified that design change notice (DCN) DM3-S-[0068 & 12621-93, sheet
5, details on the Joseph Oat Corporation fabrication drawing (no. 5659, rev. 7) for the
j
letdown heat exchanger appeared to conflict with ASME Code requirements. The drawing
notes indicated that 21 of the existing 28 heat exchanger lower flange studs were to be
j
replaced with new studs manufactured of SA-564, Gr6de 630, Condition H1025-H1100
material having a minimum yield strength of 140 ksi. The replacement studs were
fabricated of the specified material (condition H1100) and arrived on site with certified
material test report (CMTR) data demonstrating a representative yield strength greater than
149 ksi. Since PDCR MP3-90-243, Rev.1, documented an assumption that the 21 new.
studs have adequate strength to provide the structural integrity of the flanged joint,
previously served by 28 studs, the minimum yield strength data provides an engineering
value that is key to the adequacy of this design change.
However, the ASME Boiler and Pressure Vessel Code, Section ill (1983 edition, summer
1983 addenda), subsection NC, in conjunction with Table I-7.1 of the Section lli
Appendices, requires that the design allowable stress values for this replacement stud
material be based upon a minimum yield strength of 115 k;si. Therefore, despite the CMTR
data, the PDCR and its associated DCNs take credit for a stud material yield strength in
excess of that allowed by the ASME Code. The licensee documented this discrepancy in
adverse condition report (ACR) M3-96-0159. Subsequently, the licensee documented in
ACR M3-96-0465, that the PDCR/DCN allowance to use Condition H1025-H1100 material
(i.e., heat treated to a temperature between 1025 and 1100 degrees F) was also in error in
that the ASME Code, Section Ill, does not approve use of such stud material heat treated
below 1075 degrees F.
Additionally, the inspector noted that nonconformance report (NCR) 389-239, initiated in
1989 to track the letdown heat exchanger leakage and the need for repair, had been
closed in 1991. The closure was based, in part, upon a comrnitment (number 3-89-0137)
specifying an injectable leak seal repair activity that was subssquently canceled in 1993
because of changing radiological conditions. The increased area radiation levels raised
__
_.__
.
. .
.
.
..
--- .-
- . - -
_ _ -
.-
_
i
.
46
ALARA ("as low as reasonably achievable") concerns for personnel that would be involved
in the planned repairs. Automated work order (AWO) M3-91-01633 had been issued in
conjunction with PDCR MP3-90-243, documenting the belief that NCR 389-239 would
track the heat exchanger leakage until closure with the planned gasket repairs. However,
the NCR was prematurely closed, based upon a commitment that was not fulfilled.
The premature closure of NCR 389-239 also had ASME Code ramifications in that the NCR
references ASME Section XI, IWA-5250(b) requirements to evaluate boric acid corrosion on
ferritic steel components. This was accomplished in 1989 for the carbon steel studs in the
leaking lower heat exchanger flange. However, after 21 of the 28 studs were replaced
with the stainless steel materialin 1991, in accordance with PDCR MP3-90-243, there is
no evidence of a continuing licensee evaluation of the potential wastage of the seven
remaining carbon steel studs. The potential credit for the structural strength of these studs
to augment the code allowable stress values in this joint was neither quantified, nor
documented. Therefore, the ASME Section XI criteria for boric acid corrosion
considerations appear to have also been neglected with the premature closure of the QA
tracking mechanism for this continuing leakage, i.e., NCR 389-239.
c.
Conclusions
The inspector concluded that the licensee's failure to correctly translate the technical
requirements of the ASME Code, relating to the consideration and use of replacement stud
i
material, into the design details of PDCR MP3-90-243 represents an apparent violation of
l
10 CFR 50, Appendix B, Criterion Ill for Design Control. (eel 423/96-06-15) Furthermore,
in closing NCR 389-239 based upon a commitment that itself was closed without
implementation of the resulting recommendation, the licensee missed opportunities to
continue both to track the heat exchanger leakage via a quality document and to conduct
appropriate ASME Section XI (IWA-5250) " Corrective Measures". This lapse may have
i
contributed to the lack of recognition of this issue as a code violation and minimized the
l
technical concerns until raised as a material condition problem by plant personnel in June,
1996.
,
l
IV Plant SuDDort
1
(Common to Unit 1, Unit 2, and Unit 3)
R1
Radiological Protection and Chemistry Controls
R1.1
Refuelino Outaae Radioloaical Controls
a.
Inspection Scope (83750)
The inspectors reviewed radiological controls implemented during outages, including
maintaining occupational radiation exposure as low as is reasonably achievable (ALARA),
control of radiological work, and radiological housekeeping. The inspectors made frequent
tours of the radiologically controlled areas (RCA), and discussed specific radiological
controls with the unit radiation protection supervisors, ALARA coordinators and various
radiation protection technicians.
I
!
<
l
.
.
47
b.
Observations and Findinas
l
At all three units, the level of work in the radiologically controlled area was very limited at
'
the time of this inspection. With the exception of the liquid radwaste remediation project,
essentially all work at Unit 1 had ceased. At Unit 2, reactor disassembly was required for
continuation of work and was tentatively scheduled to commence in mid to late August.
At Unit 3, work in the containment dome was essentially completed, awaiting completion
of documentation and analysis.
On July 16,1996, a group of plant personnel working on the refueling floor (108'
elevation) at Unit 1 became contaminated while attempting to remove a TriNuclear vacuum
system from the spent fuel pool. The work involved removing the four filter cartridges
from the vacuum unit, then removing the vacuum unit from the spent fuel pool. While
raising the unit from the pool, a wire attached to the bottom of the unit became entangled
with several control rod blades stored along the east wall of the spent fuel pool. One of
the blades became disengaged from its hook and several other blades were moved out of
position. Six workers were contaminated, including one worker who had a small " hot"
particle located on his face. The inspector interviewed three of the workers, including the
lead health physics technician and two deconners, conducted a tour of the refuel floor and
discussed the event with unit and site radiation protection managers. The inspector also
reviewed the radiation work permit and associated radiation protection procedures involved
a this work.
The inspector's review indicated all workers were decontaminated and whole body counts
conducted. No internal contamination was detected. The inspector determined that the
licensee performed a conservative skin dose calculation for the individual who sustained a
hot particle contamination of the skin of the face. The licensee assigned a shallow skin
dose of 2.8 rem to the individual who sustained a hot particle contamination of the face
(small area near jaw). (The NRC limit for shallow exposure of the skin is 50 rem in a
calendar year.)
With the scope and level of work severely changed, ALARA planning at all three units was
very limited. Unit 1 had revised its 1996 ALARA goal to 700 person-rem, but this was
before discontinuing work. This budget also had included nine projects where so little
documentation and work scope estimates existed that the ALARA projections were
considered only accurate to within an order of magnitude. This was considered a reflection
of poor work planning. Since the start of the Unit 1 refueling outage (RFO 15) in October
1995, a total of 852 person-rem had been expended, including 501 person-rem in 1995.
For 1996, Unit 2 had revised its ALARA goal upwards to 300 person-rem to account for
the additional scope of work. The addition of work scope added an additional 250 person-
rem to the original annual goal of 50 person-rem.
Also at Unit 2, an initiative was underway to utilize senior health physics technicians as
functional coordinators for critical path evolutions (e.g., reactor disassembly / reassembly,
steam generator testing and selected valve work). Although the technicians would not
have the authority of outage coordinators, this initiative was a step in establishing
_
__
_.
. ..
___
_
. _ . .
_ . ._
.
48
1
ownership for these critical jobs. These individuals would not have collateral radiation
protection duties during their tenure as functional coordinators.
At Unit 3, continuing progress in control and minimization of leaking valves was noted.
This included continuing efforts by the health physics department to create and maintain a
valve data base, which included information on location, radiological history and
maintenance.
4
c.
Conclusions
i
The Unit 1 spent fuel pool event on July 16,1996, is currently under review by the NRC
(See Section U1.M1.1). The contaminations which resulted from this event had little
safety significance, and were properly handled by the unit her.ith physics staff.
Continued actions to address the lack of appropriate work control and work planning were
needed to allow for improvement of the ALARA programs. Actions taken at Unit 2 to
identify coordinators for critical work evolutions during the outage were considered a step
in improving work control.
R2
Status of Radiological Protection and Chemistry Facilities and Equipment
R 2.1
Unit Radioloaically Controlled Areas
a.
Insoection Scope (86750)
The inspector reviewed the status of the Unit 1 liquid radwaste remediation project. The
inspector interviewed the project leader and project members, and toured the lower levels
of the liquid radwaste facility. The inspector also conducted tours of the radiologically
controlled areas (RCAs) at all three units,
b.
Observations and Findinas
The inspector discussed the status of the Unit 1 liquid radwaste remediation project with
the project manager and the radwaste operations supervisor. Since the last inspection of
this f acility, the licensee completed removal of loose filter media and concentrates from the
floors of various cubicles, including the "A" concentrator. Non-destructive examination of
the floor drain and waste collector and test tanks had also been completed. No safety
significant dweets were found in the tanks during this testing. The tanks were also de-
scaled and flushed, while the overhead piping runs have been cleaned and painted. Piping
requiring repair, replacement or tear-out had been identified. The conterits of the filter
sludge and clean-up filter sludge tanks had also been removed.
Work remaining to be performed included removal of out-of-service equipment, including
the two filter sludge tanks, two evaporators and three concentrates tanks. Additionally,
new lighting fixtures were to be installed in the overhead, refurbishment of various pumps
and piping runs was to be undertaken, a new filter sludge tank was to installed, and a new
processing system (to replace the Ecodex filter) was to be procured. Due to budgetary
constraints, the removal of the tanks and vessels has been scheduled for 1997.
- - - . - -
- _ -
.
. - .
_ - . - - _
-
. - - -
- . - . .
- . . .
.
.
!
.
i
t
3
49
!
c.
Conclusions
1
i
The licensee continued to make progress in addressing the material and radiological
conditions in the Unit 1 liquid radwaste facility. A significant amount of work remains,
!
however, before the remediation project is completed. A significant improvement in
{
radiological and general housekeeping was also noticed in the maintenance area behind the
)
]
Unit 1 high pressure turbine.
i
j
R2.2
NUSCO Thermoluminescent Dosimetrv Laboratorv
i
j
a.
Inspection Scooe (83750)
i
{
The inspector reviewed the licensee's corporate thermoluminescent dosimetry (TLD)
laboratory. Included in this review were discussions with the laboratory manager,
i
j
assessment manager and laboratory staff.
$
b.
Observations and Findinas
!
!
The Northeast Utilities Service Company (NUSCO) dosimetry laboratory provides personal
4
TLDs to the staff at all three Northeast Utilities nuclear stations (Millstone, Connecticut
i
Yankee and Seabrook). In addition, the facility also provides TLDs for special uses, such
(
as area TLDs at Seabrook Station. Previously, the TLD laboratory was unon the direction
)
of the Radiological Assessment Branch (RAB), but has recently been placed under the
Millstone RPM. In July, a new laboratory manager was assigned from Millstone, formerly a
i
staff radiological engineer.
The laboratory is preparing for its scheduled biennial audit by the National Voluntary
,
j
Laboratory Accreditation Program (NVLAP), scheduled for September 1996. During the
!
previous NVLAP audit in 1994, a number of problems were identified, especially in the area
l
of timeliness of corrective actions. The inspector discussed with the laboratory manager
i
and quality assurance manager actions taken to address these concerns. Actions taken
included a tracking system for laboratory corrective actions and commitments, and a
j
heightened use of review teams from the stations to evaluate laboratory performance. The
inspector reviewed the most recent NVLAP sample analysis from 1995. The laboratory
'
successfully passed all eight evaluation categories, and also passed the new category IX.
.
The inspector also discussed with laboratory personnel recent concerns with area TLD
!
results from Seabrook Station. The laboratory staff's assessment of the problem identified
!
several concerns, including the laboratory's methodology for maintaining anneal date
)
records and the ability of the laboratory to process and analyze environmental-type TLDs.
]
The inspector discussed with the laboratory manager, Millstone RPM and the Director,
j
General Services these issues and the licensee's plans for laboratory improvement.
,
a
A
,
4
$
, -- , - - .
-- - , . , . . .
- , . . -
, , .
, . . _ - - -
, , - . , - - . , , . . -
- - ,
,
. - - - - - - -
.
_
_
_
_
_ . .
__ _
_ . . . _ _ _ _ .
_
_ _
._
.
.
50
1
4
c.
Conclusions
The NUSCO TLD laboratory continues to provide accurate assessment of the dose of
record for radiologically exposed personnel at the three Northeast Utilities nuclear sites.
Enhancements in laboratory operations were identified by the licensee as necessary to
maintain performance in this area.
j
'
'
R5
Staff Training and Qualification in Radiological Protection & Chemistry-
,
R 5.1
Staff Trainina and Qualification in Radioloaical Protection & Chemistry
,
1
a.
Insoection Scope (83750)
The inspector reviewed the qualifications of the interim radiation protection manager, and
discussed the health physics technician training programs with members of the technical
training staff. The inspector also reviewed new training initiatives involving plant
radiological workers,
b.
Observations t nd Findinas
Recently, the Radiation Protection Manager (RPM) for Milistone Station resigned his
position. The radiological engineering supervisor was immediately named interim RPM.
The inspector reviewed the interim RPM's qualifications and determined that the individual
met the plant technical specifications at all three units to serve in this position.
The inspector reviewed the continuing training program provided tn health physics
technicians, through review of course outlines and materials, and interviews with members
of the technical training staff. At the time of this inspection, a cycle of training on
i
Connecticut Yankee was being provided to the Millstone Station personnel, to facilitate the
loaning out of technicians during outage operations, and to assist during emergencies. The
inspector also toured a mock-up training facility being utilized by operations personnel to
enhance their performance in a radiological environment.
c.
Conclusions
The RPM position has been temporarily filled by a fully qualified health physics
professional. The technical training department continues to provide timely, in-depth
training and support to the health physics staff.
R8
Miscellaneous Radiological Protection & Chemistry issues
R8.1
Miscellaneous Radioloaical & Chemistry issues
A recent discovery of a licensee operating their facility in a manner contrary to the Updated
Final Safety Analysis Report (UFSAR) description highlighted the need for a special focused
review that compares plant practices, procedures and/or parameters to the UFSAR
descriptions. While performing the inspections discussed in this report, the inspector
reviewed the applicable portions of the UFSAR that related to the areas inspected. The
_ __
-_..
- . - . - -_. -- ..
.
.
- .
- . - . - - . - .
.
.
. _ _ _
_
.
P
51
inspector verified that the UFSAR wording was consistent with the observed plant
practices, procedures and/or parameters.
i
P3
EP Procedures and Documentation
P3.1
An in-office review of the Millstone Nuclear Power Station emergency plan
revision 21 and implementing procedure EPOP 4475, change 02, " Manager of
On site Resources (MOR)" submitted by the licensee was completed. The
inspector concluded that the revisions did not reduce the effectiveness of the
E-Plan and were acceptable.
S1
Conduct of Security and Safeguards Activities
S 1.1
Unauthorized Entrv Into the Protected Area
a.
Inspection Scope (81700)
The inspectors reviewed the event associated with an unauthorized entry into the Millstone
Station protected area (PA) by an administrative contract person.
b.
Observation and Findina
On August 5,1996, at about 8:00 a.m., an individual working for an administrative
contractor arrived at the station to report for a work assignment. The individual had
worked at the station, inside the PA until her previous assignment ended on July 19,
1996. Due to an oversight, she did not surrender her badge and key card upon termination
(under favorable conditions). However, her key card had been deactivated. The individual
assumed that her security badge would allow access to the station for the new
assignment.
The individual proceeded to the access control center, rather than the processing center
where she was directed to report. She made a telephone call to an individual, already
inside the PA, who would be a co-worker and requested to be escorted to her new work
area because she was not sure of its location. When the co-worker arrived at the access
control center, she saw that the individual was having trouble entering through the access
portal and used her own valid key card and hand geometry to allow the individual to enter.
The co-worker assumed there was a problem with the turnstile. The co-worker followed
the unauthorized individual into the PA by keying in a second time. The two individuals
reportedly worked in proximity to each other in the PA for the entire day shift.
When the individual with the deactivated key card attempted to exit the station at the end
of the shift at about 3:4 ) p.m., the deactivated key card caused an alarm to which the
security force responded. Upon questioning the individual, the unauthorized access earlier
in the day was identified.
Interviews of both individuals and a review of the computer access record by the licensee
indicated that neither individual had entered a vital area during the shift. The licensee
promptly implemented its procedure for an unauthorized individual in the PA. The licensee
,
.
O
52
initiated an investigation which is continuing. The licensee does not suspect that any
malevolence was intended and disciplinary action for both individuals is pending,
c.
Conclusions
During this event, an individual failed to comply with the licensee's requirements and
conditions of unescorted access authorization. This issue is unresolved (URI 245/96-06-
16) pending completion of the licensee's corrective actions and further NRC review.
S8
Miscellaneous Security and Safeguards issues
S8.1
General
On August 1,1994, the Commission amended 10 CFR Part 73, " Physical Protection of
Plants and Materials," to modify the design basis threat for radiological sabotage to include
the use of a land vehicle by adversaries for transporting personnel and their hand-carried
equipment to the proximity of vital areas and to include the use of a land vehicle bomb.
The amendments require reactor licensees to install vehicle control measures, including
vehicle barrier systems (VBSs), to protect against the malevolent use of a land vehicle.
Regulatory Guide 5.68 and NUREG/CR-6190 were issued in August 1994 to provide
guidance acceptable to the NRC by which the licensees could meet the requirements of the
amended regulations.
A February 29,1996, letter from the licensee to the NRC forwarded Revision 24 to its
physical security plan. The letter stated, in part, that vehicle control measures meet or
exceed all maximum parameters of design basis threat criterion and specifications found in
Regulatory Guide 5.68 and NUREG/CR-6190. A NRC July 3,1996, letter advised the
licensee that the changes submitted had been reviewed and viere determined to be
consistent with the provisions of 10 CFR 50.54(p) and were acceptable for inclusion in the
NRC-approved security plan.
This inspection, conducted on July 22 and 23,1996, in accordance with NRC Inspection
Manual Temporary Instruction 2515/132, " Malevolent Use of Vehicles and Nuclear Power
Plants," January 18,1996, assessed the implementation of the licensee's vehicle control
measures, including vehicle barrier systems, to determine if they were commensurate with
regulatory requirements and the licensee's physical security plan.
S8.2
Vehicle Barrier System (VBS)
a.
inspection Scoce
The inspectors reviewed documentation that described the VBS and physically inspected
the as-built VBS to verify it was consistent with the licensee's summary description
submitted to the NRC and was in accordance with the provisions of NUREG/CR-6190.
.
.
53
b.
Observations and Findings
l
The inspectors' walkdown of the VBS and review of the VBS summary description
disclosed that the as-built VBS was consistent with the summary description and met or
exceeded the specifications in NUREGICR-6190. During the physical inspection of the
VBS, the inspectors noted that the VBS in one area was only marginally acceptable. After
discuccion between the licensee and the inspectors, it was determined that the
effectiveness of barrier could be significantly enhanced through a sirnple modification. The
modification was completed at this location prior to the conclusion of the inspection.
c.
Conclusion
The inspectors determined that there were no discrepancies in the as-built VBS or the VBS
summary description.
S8.3
Bomb Blast Analysis
a.
Inspection Scoce
The inspectors reviewed the licensee's documentation of the bomb blast analysis and
verified actual standoff distances provided by the as-built VBS.
b.
Observations and Findinas
The inspectors' review of the licensee's documentation of the bomb blast analysis
determined that it was consistent with the summary description submitted to the NRC.
The inspectors also verified that the actual standoff distances provided by the as-built VBS
were consistent with the minimum standoff distances calculated using NUREG/CR-6190.
The standoff distances were verified by review of scaled drawings, and actual field
measurements.
c.
Conclusion
No discrepancies were noted in the documentation of bomb blast analysis or actual
standoff distances provided by the as-built VBS.
S8.4
Procedural Controls
a.
Insoection Scoce
The inspectors reviewed applicable procedures to ensure that they had been revised to
include the VBS.
b.
Observations and Findinas
The inspectors reviewed the licensee's proceduras for VBS access control measures,
surveillance and compensatory measures. The procedures contained effective controls to
provide passage through the VBS, provide adequate surveillance and inspection of the
.
.
.
.
54
VBS, and provide adequate compensation for any degradation of the VBS. The inspectors'
review of the procedure for compensatory measures disclosed that clarification of certain
requirements in the document would make it more user-friendly. Security Procedure SEP
5019, " Compensatory Measures," was revised prior to the completion of the inspection to
clarify those portions of the procedure.
The inspectors also reviewed the licensee's Common Operating Procedure C-OP-200,
" Respond to Severity Events," Rev. O, June 10,1993. This procedure provides guidance
for plant operations personnel during a security event and directs the operations personnel
into the appropriate Emergency Plan classifications. The inspectors' review of this
,
procedure disclosed that it provided guidance for situations that would include a land
vehicle bomb detonation at the VBS.
j
c.
Conclusions
The inspectors' review of the procedures applicable to the VBS disclosed no discrepancies.
V. Manaaement Meetinas
X1
Exit Meeting Summary
The inspectors presented the inspection results to members of licensee management at the
conclusion of the inspection. The licensee acknowledged the findings presented.
X1.2
Final Safety Analysis Report Review
A recent discovery of a licensee operating their facility in a manner contrary to the updated
final safety analysis report (UFSAR) description highlighted the need for additional
verification that licensees were complying with UFSAR commitments. All reactor
inspections will provide additional attention to UFSAR commitments and their incorporation
into plant practices, procedures and parameters.
While performing the inspections which are discussed in this report the inspectors
reviewed the applicable portions of the UFSAR that related to the areas inspected. The
following inconsistencies were noted between the wording of the UFSAR and the plant
practices, procedures and/or parameters observed by the inspectors as documented in
Sections U2.03.1, and U3.M3.1.
Security requirements are not specifically included in the UFSAR; they are in the licensee's
NRC-approved security plan. While performing inspections discussed in this report, the
inspectors reviewed applicable portions of regulatory requirements that related to the areas
inspected. In addition to inspecting the licensee's VBS, the inspectors also reviewed the
licensee's Protected Area Barriers (PA). The criteria for PA Barriers are contained in 10 CFR 73.2,10 CFR 73.55(c)(1) and the licensee's NRC-approved security plan. The
inspectors conducted a physical inspection of the PA Barriers (excluding the intake
_.
.
_
.
.
- _ _
.
.
55
structures) and determined that all barriers were installed and maintained as required by the
security plan and applicable regulatory requirements. No discrepancies were noted.
X3
Management Meeting Summary
XS.3
Droo-in Meetina Bv NU Manaaers
a.
Inspection Scoce (92904)
Mr. T. Harpster and Mr. F. Rothen, representing NU management conducted a drop-in
meeting with NRC staff on August 13,1996, at the Region i office. Accompanying
Messrs. Harpster and Rothen were Messrs. K. Gallen and J. Gutierrez, members of the law
firm of Morgan, Lewis & Bockius. Representing the NRC were Messrs. J. Wiggins and R.
Nimitz of the DRS staff and Mr. J. Durr of DRP. The topic covered during the meeting
involved NU presenting the results of the investigation performed by Mr. Gallen related to
the Millstone, Unit 1 radwaste facility.
b.
Observations and Findinas
The licensee's investigation concluded the following:
1.
No indication or substantive evidence suggest any NU individual intentionally
provided inaccurate communications to NRC.
2.
There were instances of miscommunications; a number of individuals that
discussed conditions with NRC inspectors had no first-hand current
information. The individuals had no actual knowledge of current status and
based their answer on dated iriformation.
3.
No NU employee saw a connection with the Nine Mile Point 1 experience;
NMP1 had floating barrels which didn't exist at Millstone.
4.
NU employees believed that NRC knew of conditions in the MSP1 tank rooms.
That limited the scope of discussions in their answers to NRC questions.
5.
NU employees thought that going into a room compromised ALARA principles;
the need to conserve dose was more important because management
expectations were to reduce collective exposures so that MSP1 would lead
'
the BWR fleet.
6.
NU employees used an " answer the question" attitude and thus did not
volunteer extra information that might have been related to the topic of the
question.
l
7.
No entries were made in the blocked-off tank rooms from about 1990 to
I
November 1994. An entry was made into one room in reaction to indications
of a leak. That November 1994 entry confirmed the existence of the leak plus
j
material on the floor. A PIR and NCR were issued; final corrective action was
j
l
. - - - .
- - - . . _
. . - - - - - . . . . - - . _ - . . _ . -
- - . - . -
-
.
. - - . -
- *
1
i
i
!
.
<
l
j
56
)
<
l
deferred. Meanwhile, personnel changes occurred along with procedure
j
changes.
'
1
l
j
8.
The PIR issued in November 1994 was closed after the planning for the
j
modification was completed; not the implementation of the modification,
~
j
9.
Problems existed in management accountability and the articulation and
'
enforcement of management expectations.
!
10.
NU committed to put their investigation report on the docket. If that report
!
contains privacy issues, NU will submit a full and a redacted report with the
j
necessary affidavit to claim the basis for withholding the full report.
I
C.
Conclusions
i,
1.
The licensee's review uncovered no evidence of NU staff intentionally
!
misleading the NRC.
2.
NU staff tended to answer NRC questions very narrowly and they also did not
have first-hand, current information when they addressed NRC questions.
l
3.
Significant problems with management accountability and managem 't
i
expectations adversely affected the radwaste systems issues. Simile
}
problems existed in other areas.
4.
Radwaste problems were identified in PIRs; the PIRs were closed based on a
4
'
plan of attack being developed and not corrective action implementation.
4
l
Other reports have addressed NRC assessment of the technical, performance, management
{
and enforcement issues.
!
i
!
I
t
i
!
!
4
i
j
j
i
i
i
!
'
- - , - - . . . -
--
,
-
-
-
..
- . - .
-
. . -
_
_ - .
. . _ _ . .
_
_ _ . .
.
__
__.
.
._
__
__ .
.
.
l
57
l
lNSPECTION PROCEDURES USED
IP 37550:
Engineering
l
IP 37551:
Onsite Engineering
IP 40500:
Licensee Self-Assessments Related to Safety issues inspections
IP 61726:
Surveillance Observations
IP 62703:
Maintenance Observations
IP 62707:
Maintenance Observations
IP 71707:
Plant Operations
IP 71750:
Plant Support Activities
IP 81700:
Physical Security Program for Power Reactors
IP 83750:
Occupational Radiation Exposure
IP 86750:
Solid Radioactive Waste Management and Transportation of Radioactive
Materials
IP 92700:
Onsite follow-up of Written reports of Nonroutine Eve nts at Power Reactor
Facilities
IP 92901:
Follow-up Operations
IP 92903:
Follow-up Engineering
IP 93702:
Prompt Onsite Response to Events at Operating Power Reactors
l
l
.
,
58
ITEMS OPENED AND CLOSED
OPEN
URI 245/96-06-01
U1.01.3 Drywell fire
URI 245/96-06-02
U1.M1.1 Spent fuel pool filter removal
URI 245/06-06-03
U1.M1.2 inservice inspection program
eel 245/96-06-04
U1.E8.1 Nonconformance Reports
eel 336/96-06-05
U2.03.1 Boric acid sampling
URI 336/96-06-06
U2.M8.2 EDG overload during surveillance
URI 336/96-06-07
U2.E1.2 Refueling pool drain line
URI 336/96-06-08
U2.E8.1 Shutdown cooling system water hammer
URI 336/96-06-09
U2.E8.2 Core thermal power exceeded
URI 336/96-06-10
U2.E8.4 Containment sump screen mesh size
eel 336/96-06-11
U2.E8.4 Failure to identify containment sump screen mesh size
eel 336/96-06-12
U2.E8.6 EEQ of solenoid valve electrical connectors
eel 423/96-06-13
U2.08.3 QSS/RSS piping outside the design basis
URI 423/96-06-14
U3.E1.1 ECCS throttle valves potential clogging
eel 423/96-06-15
U3.E8.1 Letdown heat exchanger stud material yield stress
URI 423/96-06-16
U3.S1.1 Unauthorized entry into the protected area
CLOSED
URI 245/96-05-07
U1.E8.1 Nonconformance Reports
U2.08.1 RCS Heatup rate
U2.08.2 Service water strainer backwash
"
"
"
U2.08.3 RCS Cooldown rate
U2.M8.1 RPS Surveillance
U2.M8.2 EUG Overload during surveillance
U2.M8.3 Offsite electrical alignment
U2.M8.4 Containment personnel airlock
U2.M8.5 Charging system valve leak
U2.M8.6 Missing valve internals
U2.E8.1 Failed snubbers on SDC
U2.E8.2 Core thermal power limit exceeded
U2.E8.3 Flooding of service water pumps
U2 E8.4 Containment sump screen mesh size
U2.E8.5 Wide range nuclear instruments
U2.E8.6 EEQ of solenoid operated valves
U3.08.1 Control room envelope pressurization
U3.08.2 4160V switchgear cabinet seismic qualification
U3.08.3 QSS/RSS piping outside the design basis
URI 423/96-04-12
U3.08.3 QSS/RSS piping outside the design basis
URI 423/96-05-13
U3.E8.1 Letdown heat exchanger leakage
___
_ . _ .
._ _
_ _ _ . _
_ _ _ _ _
_ __
_.
_ _ _ _ _ _ .
-
,
'
,
59
LIST OF ACRONYMS USED
,
ACR
adverse condition report
{
as low as reasonably achievable
'
ANSI /ANS American National Standards Institute /American Nuclear Society
l
abnormal operating procedure
American Society of Mechanical Engineers
AWO
automated work order
BAST
boric acid storage tank
boiling water reactor
reactor plant component cooling
control element assembly
CFR
Code of Federal Regulations
CHS
charging system
1
Configuration Management Plan
certified material test report
'
CREPS.
control room envelope pressurization system
DBDP
Design Basis Documentation Package
DCN
design change notice
Division of Reactor Projects
DRT
Discrepancy Review Team
escalated enforcement
estimated critical rod position
eel
escalated enforcement item
EEQ
electrical equipment qualification
EPOP
emergency plan operating procedures
ERT
event review team
Final Safety Analysis Report
GDC
general design criterion / criteria
GL
Generic Letter
opm
gallons per minute
HFP
hot full power
high pressure safety injection
HZP
hot zero power
IFl
inspector follow item
intergranular stress-corrosion cracking
IP
inspection procedure
inservice inspection
in-service testing
JUMA
Joint Utility Management Assessment
LCO
limiting condition for operation
LER
licensee event report
loss of normal power
. _ . _ _ _ _ _ _ _
_ . .
-
. . _
_
_
-_ _.
_ _ _ _ _ _ _ _
__
_ _ . _ _ _ . . _ - _ _ _ . _ _
60
loss of coolant accident
l
low pressure safety injection
Manager of On-site Resources
)
MRT
management review team
MTF
master tracking form
metric ton uranium
MWD
megawatt days
nonconformance report
l
NFE
Nuclear Fuels Engineering
nuclear fuel section
NGP
Nuclear Group Procedure
net positive suction head
NRC
Nuclear Regulatory Commission
NRI
no recordable indications
Nuclear Safety Advisory Letter
Nuclear Safety Information Center
Nuclear Regulation
NUSCO
Northeast Utilities Service Company
National Voluntary Laboratory Accreditation Program
'
OIR
open item report
operating license
PDCR
plant design change record
Public Document Room
plant information report
plant operation review committee
quality assurance
j
QAS
Quality and Assessment Services
quench spray system
Radiological Assessment Branch
radiologically controlled area
reactor coolant pump
refueling outage
Regulatory Guide
recirculation spray system
refueling water storage tank
standby gas treatment
shutdown cooling system
spent fuel pool
Si
safety injection
Sl&A
Safety integration and Analysis
SlH
high pressure safety injection
l
shutdown margin monitors
l
rpent resin tank
l
thermo-luminescent dosimeter
TS
technical specifications
updated final safety analysis report
"
a
v
1
61
UIR
unresolved indication report
unresolved items
vehicle barrier system
VSRT
vertical slice review team
WR-NI
wide range nuclear instrument
i
!
.